--- title: FASHN AI Help Center description: Learn how to create and manage fashion images and videos with FASHN AI. Find guides for Agent, Gallery, Elements, billing, and support. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # FASHN AI Help Center Learn how to create and manage fashion images and videos with FASHN AI. Find guides for Agent, Gallery, Elements, billing, and support. Create fashion images and videos from natural-language instructions and reference images. Use Agent to generate, edit, animate, and upscale your work, then organize the results in Gallery. --- ## Get Started Agent Create and refine images and videos through a natural conversation. Gallery Browse, filter, and manage all your images and videos. Elements Choose ready-made model photos, products, backgrounds, and other references. Guides Prompting tips, model consistency techniques, and parameter deep dives. ## Popular Pages Create and edit images Create images from instructions and references, then make changes in conversation. Create videos Create a video using only text or add motion to an image. Reference images Point Agent to an input or an earlier result in the conversation. Organize your Gallery Use Collections, Labels, Liked items, and filters to manage your work. ## Browse by Topic API Documentation Integrate FASHN into your own app or workflow. ## Start creating Ready to create your first image or video? [Open Agent](https://app.fashn.ai/agent), or follow the [Quick Start Guide](/quick-start) for a guided walkthrough. --- --- title: Quick Start Guide description: Create your first image with the FASHN AI Agent. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Quick Start Guide Create your first image with the FASHN AI Agent. Follow these steps to create your first image with Agent. ### Create Your Account Sign up or log in at app.fashn.ai with Google, GitHub, or Apple. New accounts include 10 free app credits. ### Open Agent After signing in, you land on **Home**. Start from the Agent prompt, or select **Agent** in the app sidebar. ### Create Your First Image Select **Add images** and choose a product image. Then describe the result you want in plain language. For example: > Create a full-body catalog photo of a model wearing this jacket against a clean light-gray background. Select **Generate** to begin. ### Review Your Result Agent selects the appropriate image process and adds the result to the conversation. The result is also saved to [Gallery](/using-fashn/gallery). Download the image, continue the conversation with a change, or give Agent a new request. --- ## What's Next? - Learn more about the [Agent](/using-fashn/agent) and what it can do - Choose a ready-made visual reference from [Elements](/using-fashn/elements) - Learn how to [reference images](/using-fashn/agent/reference-images) from your conversation - Read [Writing Effective Prompts](/using-fashn/guides/prompting-in-fashn/writing-effective-prompts) for more guidance - Compare [App Subscription Plans](/plans-and-pricing/app-pricing) when you need more credits - Visit [Support](/legal-and-support/support) if you need help --- --- title: Social description: Follow FASHN across social channels --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Social Follow FASHN across social channels X (Twitter)@fashn_aiInstagram@fashn.aiYouTube@fashn_aiTikTok@fashn.aiLinkedInCompany pageDiscordCommunity serverGitHubOpen source and SDKs --- --- title: Account Settings description: Manage your account information and preferences --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Account Settings Manage your account information and preferences Access your account settings to review your personal details and preferences. ## Open Account Settings Click your profile picture at the bottom-left of the app. Select Account. ## Profile Information The **Profile Information** section shows your: - Profile picture - Display name - Email address ## Changing Your Account Email Your account email is tied to the social login you use to sign in and cannot be changed directly in FASHN. If you’d like to access your workspace using a different account, or give someone else access, you can [invite that account as a collaborator](/account-management/organizations#team-members) in your organization. This lets you use the same plan and shared assets while signing in separately. Your original account remains the owner and manages billing. See [Roles and Permissions](/account-management/organizations#roles-and-permissions) for access details. For questions about moving ownership or billing to another account, contact [Support](/legal-and-support/support). To update your billing contact email, see [Billing](/account-management/billing#manage-a-web-subscription). ## Preferences Use the **Preferences** section to control: | Preference | What it controls | | --------------------------- | ----------------------------------------------------------------------- | | **Product emails** | Updates about new features, helpful guides, and special offers | | **Analytics** | Helps us understand how FASHN is used so we can improve your experience | | **Advertising measurement** | Helps us understand which campaigns introduce people to FASHN | Turn each preference on or off, then select **Save preferences**. See the [Privacy Policy](https://fashn.ai/legal/privacy-policy) for more information. ## Danger Zone ### Delete Account The **Delete account** option permanently removes your account and associated data. A confirmation appears before anything is deleted. If you subscribed through a mobile app, cancel the subscription through the App Store or Google Play before deleting your account. Account deletion is not available while a mobile subscription is active. **What will be deleted:** - All your generated images and videos - Your saved models and customizations - Your API keys - Your App and Developer API subscriptions - Account settings and preferences - All associated data and files **If you are an organization owner**, the following will also happen: - You will see the organizations you own in the deletion dialog so you can confirm what will be removed. - All assets under those organizations will be deleted. - Members will lose access to those organizations and their assets. Member individual accounts will not be deleted. If you have questions about account deletion, contact [Support](/legal-and-support/support). --- --- title: Billing description: Manage your FASHN App subscription and Developer API billing --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Billing Manage your FASHN App subscription and Developer API billing FASHN has two separate billing pages: one for your App subscription and credits, and another for Developer API subscriptions and credits. ## App Billing (`/billing`) The main Billing page manages your App plan and App credits. Only the organization owner can access **App Billing** and **Credit Usage**. ### Open App Billing Click your profile picture at the bottom-left of the app. Select Billing. ### Current Plan - Plan name and billing interval - Next plan reset date - Key features included in your plan See plan options in [Plans & Pricing → App](/plans-and-pricing/app-pricing). ### App Credits - Daily Credits (Pro/Agency only): Extra usage credits that reset daily at midnight UTC. The app consumes daily credits first. - Monthly Credits: Your plan's monthly allowance. Used after daily credits. Unused monthly credits roll over while your App subscription remains active, up to a maximum balance equal to three months of your plan's allowance. - Top-Up Credits: Additional credits you buy on demand. Used last. Learn more about how credits are consumed and renewed in [Plans & Pricing → App](/plans-and-pricing/app-pricing#monthly-credits-roll-over). ### Credit Usage In the **Credits Usage** card, select **View usage log** to open **Credit Usage**. The page shows the total credits used and number of generations for the selected period. Use the filters to choose a tool and a period: **This month**, **Last 30 days**, **Last 90 days**, or **All time**. Each entry shows its date and time, tool, relevant generation details, and credits used. ### Manage a Web Subscription Select **Manage Billing** in the plan card to open the Stripe Billing Portal. In this portal you can: - View your current subscription details and next billing date - Add, change, or remove payment methods; set a default - Update billing information, such as your email or address - Cancel your subscription Select **View Invoices** in the plan card to view and download past invoices and receipts. Subscriptions purchased through a mobile app are managed through the App Store or Google Play. See [mobile subscription details](/plans-and-pricing/app-pricing#mobile-app-subscriptions) for plan limitations and billing changes. --- ## Developer API Billing (`/billing/api`) The API Billing page is separate from App Billing and manages your Developer API subscriptions and API credits. Developer API Billing is for teams integrating FASHN into their own apps or websites. API credits are separate from App credits. ### Open API Billing Click on Developer API in the left sidebar in the app. Click on API Billing. ### Current API Plan This area shows your current billing status for the Developer API. - **If you are subscribed to a commitment tier:** you'll see your plan (tier) and reset date, your monthly credits usage, and any additional topped-up credits you have. - **If you are not subscribed to a tier:** you'll simply see your topped-up credits balance. ### Credit Auto Top-up FASHN AI can automatically top up your API credits so requests don't fail when your balance dips. **How it works:** - You set a trigger threshold (balance level) and a purchase amount (credits to add, in 100-credit increments). - When your balance reaches the threshold, the system buys that amount and charges your default saved card. **Best practices:** - Pick a threshold comfortably above your typical day of usage to avoid interruptions in your service. - We recommend setting the trigger threshold to 30% of your purchase amount. For example, if you purchase 1000 credits, set the trigger threshold to 300 credits. ### Payment Methods In this section you can add, remove, and set a default payment method for API billing. --- --- title: Organizations description: Manage your workspace, members, and shared resources --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Organizations Manage your workspace, members, and shared resources Organizations let you work as a team in FASHN. Members share your plan’s App credits and access to shared assets. Every account has a default organization, even for individual users. ## Access to Organizations You see the Organizations menu when you have an active [App subscription](/plans-and-pricing/app-pricing) or someone invites you to an existing organization. ## Open Organizations Click your profile picture at the bottom-left of the app. In the menu, locate the Organizations section. Select an organization name to switch, or click Settings to manage the current one. ## Switch Organizations Selecting another organization in the profile menu switches your working context: - All actions you take happen inside the selected organization - Credits are consumed from that organization’s balance - Assets you create are owned by that organization - The Gallery shows results created by anyone in that organization You can return to your personal organization at any time from the same menu. ## Current Organization In the **Organizations** page, the top card shows the organization you are currently working in. - Organization name (owners can rename it) - Your role (Owner or Collaborator) ## Team Members Invite teammates and manage who has access to shared resources. - Your plan includes a seats quota; the header shows seats used and how many remain - Owners can click Invite to send an email invite - Each collaborator entry shows name, email, and role; owners can remove members Only owners can invite or remove collaborators. If you need to join an organization, ask an owner. Seat availability depends on your plan; owners can manage seats in [Billing](/account-management/billing). Seats are included with your App plan. See seat quantities on [Plans & Pricing → App](/plans-and-pricing/app-pricing). ## What’s Shared in an Organization - App subscription benefits and App credits - Assets your team creates under the organization ## Roles and Permissions - Owner: invite/remove collaborators, view [Billing](/account-management/billing) and manage the plan, rename the organization. - Collaborator: use shared resources and create assets within the organization. ## Danger Zone Leaving an organization removes your membership immediately. Your personal account stays active, but: - You lose access to the organization’s assets and plan - Items you created under the organization remain with the organization - You will no longer see the organization in your switcher At this time, organization owners cannot leave their organization. If you need to transfer ownership to another member, please contact [support](/legal-and-support/support). If you intend to close your personal account, review how owning an organization affects deletion in [Account Settings](/account-management/account-settings). --- --- title: Legal & Support description: Find help and review policies. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Legal & Support Find help and review policies. - [Support](/legal-and-support/support) - [Privacy Policy](https://fashn.ai/legal/privacy-policy) - [Terms of Service](https://fashn.ai/legal/terms-of-service) --- --- title: Support description: How to get help with FASHN --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Support How to get help with FASHN ## How to Get Help ### Chat with us After you complete a payment, chat support is available and is the fastest way to reach the team. - **Web:** Select **Help** in the sidebar, then select **Chat with us**. - **Mobile app:** Open the account menu, then select **Support**. Before chat support becomes available, use Discord or email. ### Other support channels - **Discord:** Join [our server](https://discord.gg/MCs39Gf4yn) and post in #support if you need help with the app or API. For product-specific discussions such as feedback or feature requests, use #web-app or #api. - **Email:** [support@fashn.ai](mailto:support@fashn.ai) - **Slack Connect (enterprise):** We can open a shared Slack channel for direct, dedicated support for your team. ## Share a Generation with Support ### Web App For a result in **Gallery** on the web: Right-click the result. Select Copy Generation ID. Paste the generation ID into your support chat and describe what happened. The generation ID lets our team review the inputs and result and see if we can offer tips to avoid the issue. ### Developer API Share the prediction ID returned as `id` by `POST /v1/run`. This is the same ID used to poll the prediction status. Include what happened and any error message. If the request failed before an ID was returned, include the endpoint, the approximate request time, and the error response. These details help us locate the request and investigate. --- --- title: API On-Demand & Commitment Tiers description: On-demand credits with optional commitment tiers for discounts and monthly included usage. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # API On-Demand & Commitment Tiers On-demand credits with optional commitment tiers for discounts and monthly included usage. FASHN API uses simple, credit-based pricing. You can start with on-demand credits and, when you're ready, add a monthly commitment tier to unlock included monthly credits and discounted top-up rates. All billing and usage are managed in the app's Developer API dashboard. --- ## On-Demand Pricing (Flexible) On-Demand$0.075 per creditAccess to API DashboardSee request history and monitor usageStrict data retention and privacy policiesPay for credits in advanceGet automated top-up alertsAvailable for commercial useMinimum purchase of $7.50 (100 credits) --- ## Credit Costs Credit costs depend on the endpoint, resolution, and generation mode you select. Pricing is charged per successful output. Failed predictions do not consume credits. ### Image Generation Endpoints These costs apply to Product-to-Model, Model Creation, Model Swap, Face to Model, Edit, Reframe, and Try-On Max: | Generation Mode | 1K | 2K | 4K | | --------------- | -- | -- | -- | | Fast | 1 | 2 | 3 | | Balanced | 2 | 3 | 4 | | Quality | 3 | 4 | 5 | If you do not select a generation mode, the system selects one automatically based on factors like the requested resolution and endpoint. Available generation modes vary by endpoint; see each endpoint's documentation for details. Additional pricing rules: - **Face Reference**: +3 credits per output image (available on Product-to-Model, Model Creation, Model Swap) - **Multiple outputs**: the per-output cost is multiplied by the number of images requested ### Fixed-Cost Endpoints | Endpoint | Credits per output | | --------------------- | ------------------ | | Virtual Try-On (v1.6) | 1 | | Background Remove | 1 | ### Video Generation | Duration \ Resolution | 480p | 720p | 1080p | | --------------------- | ---- | ---- | ----- | | 5 seconds | 1 | 3 | 6 | | 10 seconds | 2 | 6 | 12 | --- ## Commitment Tiers (Subscription) Commitment tiers include a monthly credit allowance and unlock discounted top-up rates. The included credits are a starting point, not a cap. You can top up additional credits at any time and they will be charged at the same discounted rate as your tier. The same credit costs above apply regardless of tier; the subscription only affects how much you pay per credit. If you cancel, any remaining discounted credits expire at the end of the current billing period. Tier I$19 / month + top-upsEverything in On-Demand282 monthly credits includedTop-ups: $0.0675 per credit (10% off)Tier II$249 / month + top-upsEverything in On-Demand4,150 monthly credits includedTop-ups: $0.0600 per credit (20% off)Tier III$1,249 / month + top-upsEverything in On-Demand25,594 monthly credits includedTop-ups: $0.0488 per credit (35% off) Annual billing is priced at 10 months instead of 12: | Tier | Monthly | Annual | | -------- | ---------- | ----------- | | Tier I | $19/mo | $190/yr | | Tier II | $249/mo | $2,490/yr | | Tier III | $1,249/mo | $12,490/yr | --- ## Top-Ups Purchase additional credits via a one-time Stripe checkout. Credits activate immediately. - Minimum purchase: 100 credits ($7.50 at on-demand rates), in increments of 100. - Tier discounts apply automatically while subscribed (Tier I: $0.0675, Tier II: $0.0600, Tier III: $0.0488 per credit). - On-demand top-ups expire 12 months after purchase. Discounted top-ups expire when your subscription ends. - API credits are separate from app credits. **Where to top up:** **Sidebar** → **API** → **API Billing** → **Top up API credits**. You can also use the **Top up** button in the credits bar at the top of any API page. No subscription required. **Auto top-up:** **Sidebar** → **API** → **API Billing** → **Auto top-up**. Set a balance threshold and purchase amount. Requires a saved payment method. --- ## Managing Your Plan ### Upgrading your tier **Monthly upgrades:** - Take effect immediately at the new tier's price and discount rate. - Unused credits from your current tier roll into the new tier's allowance. - Billing cycle resets to the upgrade date. **Annual upgrades:** - Prorated: the unused portion of your current plan reduces the upgrade cost. - Billing cycle and renewal date stay the same. ### Downgrading or cancelling Downgrades and cancellations take effect at the end of the current billing cycle. You retain access to your current tier and discount rate until the next renewal date. **Discounted credits purchased under a higher tier expire** at the end of the current billing cycle. ### Switching from monthly to annual There is no one-click conversion. Cancel the monthly commitment, let the cycle end, then subscribe to the annual tier you want. --- ## Refund Policy Unused API credit purchases, whether part of a commitment tier or individual top-ups, are non-refundable. If you are unsure about your expected usage, start with on-demand and top up as little as 100 credits ($7.50) at a time. --- --- title: App Subscription Plans description: Plans and pricing for the FASHN app --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # App Subscription Plans Plans and pricing for the FASHN app The FASHN app is subscription-based. Each plan gives you a monthly allowance of credits for AI tools like Product to Model, Try-On, editing, upscaling, and video generation. ## Plans Basic$19 / month200 monthly creditsUnused monthly credits roll over, up to 600 Core AI fashion tools, including Product to Model, Try-On, Packshot, Model Swap, AI Fashion Models, and Edit Up to 4K image generation and upscaling Growing library of ready-to-use AI fashion models and backgrounds AI videos up to 720pInvite 2 team membersUp to 3 simultaneous generationsPro$49 / monthEverything in Basic750 monthly creditsUnused monthly credits roll over, up to 2,25050 daily creditsMost realistic AI image generation capabilitiesAI videos up to 1080pInvite 5 team membersUp to 6 simultaneous generationsAgency$99 / monthEverything in Pro1,500 monthly creditsUnused monthly credits roll over, up to 4,500100 daily creditsRadar (AI visibility)Invite 10 team membersUp to 11 simultaneous generationsAgency 2×$199 / monthEverything in Agency3,000 monthly creditsUnused monthly credits roll over, up to 9,000200 daily credits In the app, Agency 2× appears as a "2× Credits" toggle on the Agency pricing card. Annual billing is priced at 10 months instead of 12: | Plan | Monthly | Annual | | --------- | ------- | --------- | | Basic | $19/mo | $190/yr | | Pro | $49/mo | $490/yr | | Agency | $99/mo | $990/yr | | Agency 2× | $199/mo | $1,990/yr | ### Mobile App Subscriptions Subscriptions purchased through a mobile app are managed by the app store used for the purchase. Mobile app billing is monthly only and does not currently support Agency 2× or [credit top-ups](#top-ups). Mobile app billing also affects operations such as [plan upgrades](#managing-your-plan) and [refund requests](#refund-policy), which are covered later on this page. --- ## Credit Costs Credit costs depend on the resolution and generation mode you select. If you request multiple outputs in a single run, the cost is multiplied by the number of outputs. ### Image Generation Tools These costs apply to Product to Model, Model Swap, Create Model, Edit, Reframe, Try-On, and Packshot: | Generation Mode | 1K | 2K | 4K | | --------------- | --- | --- | --- | | Fast | 1 | 1 | 1 | | Balanced | 1 | 1 | 2 | | Quality | 1 | 2 | 3 | If you do not select a generation mode, the system selects one automatically based on factors like the requested resolution and tool. ### Modifiers - **Face Reference**: +2 credits per image in the Studio (available on Product to Model, Model Swap, Create Model). The FASHN Agent adds no Face Reference surcharge. - **Multiple outputs**: cost multiplied by the number of images requested ### Fixed-Cost Tools | Feature | Cost | | ----------------- | ---- | | Face Swap | 2 | | Background Remove | 1 | | Upscale | 1 | ### Video Generation | Duration \ Resolution | 480p | 720p | 1080p | | --------------------- | ---- | ---- | ----- | | 5 seconds | 1 | 3 | 6 | | 10 seconds | 2 | 6 | 12 | - **Audio**: a video with audio costs twice the credits shown above. Audio is available when you [create videos with Agent](/using-fashn/agent/create-videos#add-audio). **Examples:** - **Product to Model** at 1K with Face Reference, requesting 2 images: (1 + 2) x 2 = 6 credits. - **Model Swap** at 2K Quality for 4 images: 2 x 4 = 8 credits. - **Video Generation** at 720p for 10 seconds: 6 credits. - **Video Generation** at 720p for 10 seconds with audio: 6 x 2 = 12 credits. --- ## Daily Credits Pro and Agency plans include daily credits that reset every 24 hours. Daily credits give you room to experiment, make quick edits, or try new ideas without cutting into the monthly credits you may want to reserve for larger projects. When you use a feature, FASHN consumes **daily credits first**, then **monthly credits**, and finally any **top-up credits** you have purchased. --- ## Monthly Credits Roll Over Unused monthly credits roll over to the next month while your App subscription remains active. Your monthly credit balance can hold up to three times your plan's monthly allowance: | Plan | Monthly Allowance | Maximum Balance | | --------- | ----------------- | --------------- | | Basic | 200 | 600 | | Pro | 750 | 2,250 | | Agency | 1,500 | 4,500 | | Agency 2× | 3,000 | 9,000 | At each monthly renewal, FASHN keeps your unused monthly credits and adds as much of the new monthly allowance as will fit within your plan's maximum balance. For example, if a Basic plan renews with 500 monthly credits remaining, 100 credits are added to bring the balance to the 600-credit limit. This limit applies only to the monthly credit balance. Daily credits continue to reset every 24 hours, and top-up credits remain separate with their own expiration date. --- ## Top-Ups Buy extra credits whenever you need more room to generate. Top-up credits are added to your account immediately after purchase. - Price: $0.10 per credit ($10 per 100 credits). - Minimum purchase: 100 credits ($10), in increments of 100. - Top-up credits expire 12 months after purchase. - App credits are separate from API credits. **Where to top up:** **Sidebar** → **Top up** in the App Credits panel. Or go to **Billing** → **App** tab → **Top up**. --- ## Managing Your Plan ### Upgrading your plan **Monthly upgrades:** - Take effect immediately at the new plan's price. - Your new monthly credit balance is set to the new plan's monthly allowance. - Unused monthly credits from your previous plan move to your top-up balance and expire after 12 months. - Billing cycle resets to the upgrade date. **Annual tier upgrades:** - Prorated: the unused portion of your current plan reduces the upgrade cost. - Billing cycle and renewal date stay the same. - Your previous monthly credit balance is replaced by the new plan's monthly allowance. Use any remaining monthly credits before upgrading. **Mobile app upgrades:** - Your app store handles the billing adjustment for the price difference between plans and the time remaining in your current billing cycle. - The upgrade takes effect immediately. - FASHN grants upgrade credits based on the amount actually paid for the new plan after the app store's adjustment. - After that first partial month, your plan renews normally and you receive the full monthly credit allowance. ### Downgrading or cancelling Downgrades and cancellations take effect at the end of your current billing cycle. You continue using your existing plan until the next renewal date. - **Downgrade:** you are charged for the new plan and switched automatically. Your monthly credit balance is capped at the new plan's three-month limit when the downgrade takes effect. Credits above that limit expire; credits at or below it remain available. - **Cancel:** the subscription ends and no new charge is made. Any remaining monthly credits expire when the plan ends. ### Switching from monthly to annual On the web, go to **Billing** → **App** and select **Switch to yearly** under your current plan's price. Review the change, then select **Schedule yearly billing**. Your current monthly plan, features, and credits remain active until the end of the billing period. The annual plan starts on your next billing date, when you are charged the full annual price. You do not need to cancel and resubscribe. The scheduled switch applies to your current plan. To choose a different plan, cancel the scheduled yearly switch first, then change plans separately. Mobile app subscriptions do not currently offer annual billing. To move from a mobile monthly subscription to an annual subscription, cancel it through your app store, wait for the current billing period to finish, then start the annual plan on the web. --- ## Refund Policy FASHN does not offer refunds for unused credits, change-of-mind purchases, or results that do not meet personal expectations. Every account includes 10 complimentary credits so you can try the core tools before paying. For subscriptions purchased through a mobile app, the app store manages payments, invoices, refunds, and billing disputes. If you need to report an unauthorized charge, suspected fraud, or another billing issue, request a refund through [Apple](https://support.apple.com/en-us/118223) or [Google Play](https://support.google.com/googleplay/answer/15574897?hl=en). --- --- title: Plans & Pricing description: Learn about FASHN plans and pricing --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Plans & Pricing Learn about FASHN plans and pricing - [App Subscription Plans](/plans-and-pricing/app-pricing) - [API On-Demand & Commitment Tiers](/plans-and-pricing/api-pricing) --- --- title: Elements description: Use curated images as references across FASHN. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Elements Use curated images as references across FASHN. Elements is a library of curated images you can use as references. Choose an Element, then describe the result you want. ## Open Elements Select **Elements** in the app navigation. Browse the full library, or choose a category to narrow the results. ## Choose the right type of Element | Category | Use it for | | --- | --- | | **Models** | Choosing the person and overall composition for an on-model image | | **Portraits** | Guiding a person's identity and facial appearance | | **Poses** | Showing the body position or gesture you want | | **Backgrounds** | Setting the location, surface, or surrounding scene | | **Photography Styles** | Guiding lighting, camera treatment, and overall visual style | | **Clothing, Accessories, and Shoes** | Using a sample product or guiding a Packshot style, such as flat-lay or ghost mannequin | ## Start from the Elements library 1. Browse **All Elements** or choose a category. 2. Select an Element. FASHN opens a new Agent conversation with the Element attached. 3. Describe the result you want, and add other images when needed. For example, select a Model Element, attach your own product image, and ask: > Create a clean catalog image of this model wearing the attached linen dress against a warm white studio background. ## Choose an Element in Agent Open **Add images** in the Agent message box and select **Browse Elements**. Choose a category, then select an image to add it to your request. You can combine Elements with uploads, [Gallery](/using-fashn/gallery) images, and earlier results. Use [image references](/using-fashn/agent/reference-images) to point Agent to an exact input or result. ## Use Elements in Studio Studio parameters such as **Face Reference**, **Background Reference**, and **Image Prompt** include an **Elements** option. Each opens on the most relevant category, but you can switch categories. For example, open **Elements** from the model image field in Try-On to choose a model: ## Access Every paid App plan includes the full Elements library. ## What's Next? - Use Elements while [creating or editing images with Agent](/using-fashn/agent/create-and-edit-images). - Learn how to [reference an exact image](/using-fashn/agent/reference-images) in a conversation. - Browse and organize your own work in [Gallery](/using-fashn/gallery). --- --- title: Credits for bad results description: Why bad results still use credits, and when to contact support about a result or charge. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Credits for bad results Why bad results still use credits, and when to contact support about a result or charge. ## When a Result Misses AI generation is not fully predictable, for images and videos alike. A model can miss an instruction or add details you did not ask for, even with clear inputs and a well-written prompt, and the same request can turn out differently from one generation to the next. When this happens, it does not mean you set up your request incorrectly or that something is wrong with the service. We know it is frustrating to spend credits on a result you cannot use. Every generation runs the full computation needed to create the output, and that processing has the same cost whether the result is right or misses a detail. Credits pay for each generation, so we cannot charge only for results that match your request. A retry is a new generation, so it uses credits too. Credits used on completed generations are not refunded or restored. See the [App refund policy](/plans-and-pricing/app-pricing#refund-policy) for details. To get better results in fewer tries, browse the [Guides](/using-fashn/guides) or watch the tutorials on our [YouTube channel](https://www.youtube.com/@fashn_ai). ## When to Contact Support Contact [Support](/legal-and-support/support) so the team can investigate if: - A generation ended with an error or an output is missing. - You see a credit charge you do not recognize. If a completed result keeps missing in the same way, [share its generation ID](/legal-and-support/support#share-a-generation-with-support). The team can review the inputs and result and suggest what to change on your next try. --- --- title: Enable hardware acceleration in your browser description: Steps to enable WebGL and hardware acceleration so the built-in image editor in FASHN can load. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Enable hardware acceleration in your browser Steps to enable WebGL and hardware acceleration so the built-in image editor in FASHN can load. FASHN's built-in image editor (the crop, rotate, and annotate tool you open from the Studio) uses **WebGL** to render the canvas. WebGL relies on **hardware acceleration**, which lets your browser draw graphics with your GPU instead of your CPU. If you see the message **"Missing WebGL Support"** when opening the editor, your browser most likely has hardware acceleration turned off, or your device is using an outdated browser or graphics driver. This guide walks you through the fix. These settings live in your browser, not in FASHN. After changing them you will need to restart the browser and reopen the editor. ## Before changing any settings Try the quick checks first. They resolve most cases without touching browser flags. ### Update your browser Open your browser's **About** page and install any pending update. Older versions of Chrome, Edge, Firefox, and Safari ship with stricter graphics policies that can block WebGL on some hardware. ### Restart your computer A reboot clears stuck GPU processes and applies any pending graphics driver updates. ### Disable extensions that block scripts or graphics Privacy and script-blocking extensions (uBlock Origin in strict mode, NoScript, Privacy Badger, and similar) can disable WebGL on individual sites. Pause the extension on `app.fashn.ai` and reload. If the editor still shows the missing WebGL message, enable hardware acceleration using the steps for your browser below. ## Enable hardware acceleration ### Google Chrome Click the three-dot menu in the top-right corner, then choose **Settings**. In the sidebar, open **System**. Turn **Use hardware acceleration when available** on. Click **Relaunch** to restart Chrome. To confirm WebGL is working, open a new tab and go to `chrome://gpu`. Look for **WebGL** and **WebGL2** under "Graphics Feature Status". They should read *Hardware accelerated*. ### Microsoft Edge Click the three-dot menu in the top-right corner, then choose **Settings**. In the sidebar, open **System and performance**.Turn **Use hardware acceleration when available** on.Click **Restart** to apply the change. To confirm, open a new tab and go to `edge://gpu`. **WebGL** and **WebGL2** should read *Hardware accelerated*. ### Mozilla Firefox Click the menu button (three lines) in the top-right corner, then choose **Settings**. Stay on the **General** tab and scroll down to **Performance**. Uncheck **Use recommended performance settings**, then check **Use hardware acceleration when available**. Restart Firefox. To confirm, open a new tab, go to `about:support`, and check **Graphics**. Look for **WebGL 1 Driver Renderer** and **WebGL 2 Driver Renderer**. They should list your GPU, not a software fallback. If WebGL was explicitly disabled in Firefox, you may also need to open `about:config` and confirm that `webgl.disabled` is set to `false`. Only change this setting if you (or an extension) previously turned it on. ### Safari (macOS) Hardware acceleration is on by default in Safari, so the missing WebGL message usually points at an experimental feature or an old macOS build. Update macOS to the latest version available for your Mac (Apple menu, **System Settings**, **General**, **Software Update**). In Safari, open **Safari** in the menu bar, choose **Settings**, then go to **Advanced**. Enable **Show features for web developers** (in older Safari versions, this is **Show Develop menu in menu bar**). From the menu bar, choose **Develop**, **Developer settings**, **Experimental Features**, and make sure **WebGL via Metal** is enabled. Do not disable any WebGL-related feature here. Restart Safari and reopen the editor. Safari on iPad and iPhone supports WebGL out of the box on recent iOS versions. If the editor still does not load on iOS, update iOS to the latest version, or open FASHN in a desktop browser. ### Brave, Opera, Vivaldi, and other Chromium browsers These browsers share Chrome's underlying engine, so the steps are identical to **Google Chrome** above. Open **Settings**, **System**, and turn on **Use hardware acceleration when available**, then relaunch. For Brave specifically, also check **Settings**, **Shields**, and make sure Shields is not blocking scripts on `app.fashn.ai`. ## Still not working? If hardware acceleration is on and you have updated your browser, your device may not support WebGL with its current setup. Try the following in order: - **Switch browsers.** Open FASHN in Chrome, Edge, or Firefox on a desktop computer. - **Update your graphics drivers.** On Windows, run Windows Update or use your GPU vendor's tool (NVIDIA GeForce Experience, AMD Software, Intel Driver & Support Assistant). On macOS, drivers ship with the OS, so a system update is enough. - **Try a different device.** If you are on a managed work laptop, GPU drivers and WebGL may be restricted by your IT policy. - **Skip the editor for now.** You can still upload, generate, and download images without entering the built-in editor. Use a desktop image editor for crops and annotations until WebGL is available. If none of the above works, reach out through any of the channels listed on the [Support](/legal-and-support/support) page and include: - Your browser name and version. - Your operating system and version. - A screenshot of the **Missing WebGL Support** message. - (Optional, but very helpful) The contents of `chrome://gpu`, `edge://gpu`, or the Graphics section of `about:support` in Firefox. --- --- title: Troubleshooting description: Guides for generation, credit, browser, device, and configuration issues in FASHN. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Troubleshooting Guides for generation, credit, browser, device, and configuration issues in FASHN. Guides for resolving generation, credit, browser, device, and configuration issues in FASHN. If you cannot find your issue here, reach out through the [Support](/legal-and-support/support) page. --- --- title: Create and edit images description: Create new images, work with products and models, and refine results with Agent. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Create and edit images Create new images, work with products and models, and refine results with Agent. Agent can create an image using only text, work from one or more images, or refine an earlier result in the same conversation. ## Create an image using only text Describe the image you want, including the subject, setting, composition, and style that matter to you. An image is not required. For example: > Create a full-body catalog photo of a model wearing a plain white T-shirt and straight-leg jeans, standing front-facing against a clean light-gray background with even lighting. You can also create a model by describing traits such as appearance, outfit, pose, framing, and setting. ## Create from an image Attach a product, model, or other source image and describe the result you want. Agent can: - Generate a new model wearing a product. - Put clothing on a model in an existing image. - Replace the person in an existing image while keeping the outfit, pose, and setting. - Create a product-only image for a catalog or online store. - Edit an image by changing its background, colors, pose, styling, or framing. For example: > Create a full-body campaign photo of a model wearing this jacket with black trousers, walking through a modern city setting. ❌ Avoid adding preservation instructions: > Put this jacket on a model and preserve the embroidery, stitching, colors, and every product detail exactly. Agent already accounts for visible product details and parts of the image outside the requested change. Focus your request on the result you want. ## Work with multiple images You can use up to four images in one request. Identify what each image should contribute when their roles are not obvious, such as the product, model, background, or visual style. For example: > Put the shirt from the first image on the model in the second image, using the third image as the setting. Use [image references](/using-fashn/agent/reference-images) to select exact images from the current conversation. ## Edit an image or result Attach or reference the image you want to change, then describe the change. Continue the conversation to refine the result without starting again. For example: - “Change the background to a hotel lobby at night.” - “Make the jacket navy blue.” - “Use a closer crop and have the model face the camera.” - “Create another variation with softer lighting.” When several images are available, [reference the exact result](/using-fashn/agent/reference-images) you want to edit. ## Change the framing or aspect ratio Ask Agent to make an existing image square, portrait, landscape, or a specific aspect ratio. Agent can crop the image or expand the surrounding scene to fit the new frame. For example: > Make this image 16:9 and expand the background on both sides. For a new image, you can also choose the aspect ratio before generating. See [Generation settings](/using-fashn/agent/generation-settings). ## What's Next? - Learn how to [reference images](/using-fashn/agent/reference-images) from the current conversation. - [Create a video](/using-fashn/agent/create-videos) from text or an image. - [Upscale your final image](/using-fashn/agent/upscale-images) to increase its resolution. --- --- title: Create videos description: Create a video using only text, animate an image, or guide the ending with a second image. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Create videos Create a video using only text, animate an image, or guide the ending with a second image. Agent can create a video using only text, animate an image, or use two images to guide how a video begins and ends. ## Create a video using only text Describe the subject, setting, action, and camera movement. Agent will create the video without a start image. For example: > Create a 5-second cinematic video of a silver handbag on a rotating pedestal in a dark showroom. Use a slow camera push-in, soft moving reflections, and one continuous shot. ## Animate an image Attach or [reference](/using-fashn/agent/reference-images) the image you want to animate, then describe the motion. The image establishes the subject, product, setting, and starting composition. Focus your request on how the subject and camera should move, and what should remain still. For example: > Turn this image into a 5-second video. The model slowly turns to show the back of the jacket while the camera moves in a gentle arc. ### Change the image before animating it If the start image also needs a change, include the new visual result and the motion in the same request. Agent will prepare the image first, then create the video. For example: > Place this handbag on a marble pedestal in a dark showroom, then create a 5-second video with a slow camera push-in and moving reflections. ## Use a start image and an end image Use two images when you want more control over how the video begins and ends. The **start image** sets how the video begins, while the **end image** guides how it ends. Attach or reference both images and identify their roles in your request. For example: > Begin with the first image and end with the second image. Over 5 seconds, the model turns naturally while the camera moves from the front view to the final side view. Choose images that can be connected through a believable camera movement or subject action. Keep the same main product or subject visible across both images. ## Guide the motion Keep the request focused on one continuous shot. Describe: - How the subject or product moves. - Whether the camera stays still, pushes in, pans, or moves around the subject. - Which parts of the scene should remain still. - The final position or composition when it matters. Use [a start image and an end image](#use-a-start-image-and-an-end-image) when you need more control over how the video ends. ## Add audio Videos are silent unless you ask for sound. To add audio, describe what should be heard in the same request: speech, music, ambience, or the sounds of the action. For example: > Create a 5-second video of the model walking toward the camera through the concrete hall. Add the sound of her footsteps on the polished floor. - **Speech:** Write the exact words in quotation marks. Speech works best in English. - **General sound:** If you ask for sound without describing it, Agent adds subtle ambience and the natural sounds of the action. - **Silent video:** Ask for a silent video at any time to turn audio off again. A video with audio costs twice the credits of the same video without audio. See [video credit costs](/plans-and-pricing/app-pricing#video-generation). ## Choose the duration and resolution Agent supports the following video options, subject to your plan: | Option | Available values | How to choose | | --- | --- | --- | | Duration | 5 or 10 seconds | Include the duration in your request. Agent uses 5 seconds when you do not specify one. | | Resolution | 480p, 720p, or 1080p | Include the resolution in your request, or use Generation mode in the settings menu. | | Audio | Off or on | Off unless you ask for sound. See [Add audio](#add-audio). | When Generation mode is selected, **Fast** uses 480p, **Balanced** uses 720p, and **Quality** uses 1080p. **Auto** lets Agent choose. See [Generation settings](/using-fashn/agent/generation-settings). ## Limitations - Existing videos cannot be used as inputs. - Generated videos cannot be used as image references in later requests. - Videos cannot be transparent. Transparent areas in a start image or an end image are filled with solid black or white. ## What's Next? - Learn how to [reference a start image and an end image](/using-fashn/agent/reference-images). - [Create or edit a start image](/using-fashn/agent/create-and-edit-images) before adding motion. - Review [generation settings](/using-fashn/agent/generation-settings) for video quality and resolution. --- --- title: Generation settings description: Choose the aspect ratio, resolution, speed, and quality for Agent generations. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Generation settings Choose the aspect ratio, resolution, speed, and quality for Agent generations. ## Open Generation settings Select the gear icon (⚙️) beside the send button in Agent. The same menu is available when you open Agent from Gallery. **Auto** is a good default for most requests. Choose a specific setting when you need a particular shape, image size, generation speed, or quality level. ## Aspect ratio Aspect ratio controls the shape of a generated or edited image. Choose **Auto** to let Agent select a suitable ratio, or choose one of the available values: `1:1`, `2:3`, `3:2`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, or `21:9`. Common choices include: | Ratio | Common fashion and commerce uses | | --- | --- | | 1:1 | Square product images, marketplace listings, and product grids | | 2:3 | Full-body on-model images and catalog photography | | 4:5 | Portrait social feed posts and product campaigns | | 9:16 | Instagram Stories and Reels, TikTok, and vertical ads | | 16:9 | Website banners and landscape campaign assets | | 21:9 | Wide storefront hero images | To change the aspect ratio of an existing image, choose the target ratio and ask Agent to crop or expand the image to fit it. See [Create and edit images](/using-fashn/agent/create-and-edit-images#change-the-framing-or-aspect-ratio). ## Image resolution Resolution controls the output size of new images and edits. | Setting | Approximate size | When to use | | --- | --- | --- | | Auto | Selected by Agent | Most requests | | 1K | ~1MP | Quick drafts and smaller outputs | | 2K | ~4MP | Most image creation and editing | | 4K | ~16MP | Highest output tier for final assets | Exact dimensions depend on the image's aspect ratio. For most workflows, create and edit at 2K, then [upscale only your selected final results](/using-fashn/agent/upscale-images#recommended-workflow) to 4K. Working at 2K is faster and more cost-efficient, and generally makes complex edits easier to control, with results that follow your instructions more reliably. ## Generation mode Generation mode controls the balance between speed and quality. | Mode | Behavior | | --- | --- | | Auto | Selected by Agent. | | Fast | Prioritizes shorter generation times. | | Balanced | Balances speed and output quality. | | Quality | Prioritizes higher image quality and may take longer. | For video, Generation mode also determines resolution: | Mode | Video resolution | | --- | --- | | Auto | Selected by Agent. | | Fast | 480p | | Balanced | 720p | | Quality | 1080p | Quality mode is available on Pro and Agency plans. Credit use varies by mode. See [App Subscription Plans](/plans-and-pricing/app-pricing) for current access and costs. ## Apply your settings On desktop, each selection takes effect immediately. On mobile, choose your values and select **Apply**. Active settings are included automatically with each request, so you do not need to repeat them in your instructions. ## What's Next? - Apply the settings while [creating or editing images](/using-fashn/agent/create-and-edit-images). - See how Generation mode affects [video creation](/using-fashn/agent/create-videos#choose-the-duration-and-resolution). - Learn when to [upscale an existing image](/using-fashn/agent/upscale-images). --- --- title: Agent description: Create and refine images and videos in conversation with the FASHN AI Agent. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Agent Create and refine images and videos in conversation with the FASHN AI Agent. The **Agent** helps you create images and videos through a natural-language conversation. Describe what you want in your own words, add images when helpful, and continue with new instructions to refine a result or create something completely new. ## Open the Agent ### From Home or the app sidebar Select **Agent** in the app sidebar, or enter a prompt from Home. You can start a new conversation or reopen an earlier one from the conversation history. ### From Gallery Agent is also available at the bottom of Gallery. Enter a request in the Agent message box, or add an image from Gallery when you want to use it in your request. ## How it works 1. **Describe the result:** Name the subject, product, setting, or change you want in plain language. 2. **Add images:** Upload images, choose them from Gallery or [Elements](/using-fashn/elements), or [reference an image](/using-fashn/agent/reference-images) from the current conversation. 3. **Generate:** Agent chooses the appropriate image or video process and creates the result. 4. **Continue the conversation:** Ask for a change, refer to an earlier image, or create a variation. Start a new conversation when you begin an unrelated workflow. Generated images and videos appear in the conversation and are also saved to [Gallery](/using-fashn/gallery). ## What you can do ## Limitations - Existing videos cannot be used as inputs. - Images cannot be fetched from external URLs. - Account and billing changes are not supported. ## What's Next? - Learn how to [create and edit images](/using-fashn/agent/create-and-edit-images). - Use [image references](/using-fashn/agent/reference-images) to identify a specific input or earlier result. - Learn how to [create videos](/using-fashn/agent/create-videos), including how to use a start image and an end image. - [Upscale an image](/using-fashn/agent/upscale-images) to increase its resolution. - Create [transparent images](/using-fashn/agent/transparent-images) or remove a background. - Choose your [generation settings](/using-fashn/agent/generation-settings) for aspect ratio, resolution, speed, and quality. --- --- title: Reference images description: Use image references to identify the exact inputs and results Agent should use. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Reference images Use image references to identify the exact inputs and results Agent should use. An `@` reference identifies a specific image in the current conversation. References are available for uploaded images, images selected from Gallery or [Elements](/using-fashn/elements), and images generated earlier in the conversation. ## Reference an image directly Hover over an attached input or an image in **Media history**, then select its `@` button. The image is added to the message box as a reference chip. Generated image results also include a **Reference** button below the result. Selecting an image in Media history takes you to its position in the conversation. Selecting its `@` button adds the image to the next message. ## Choose an image from the message box Type `@` in the message box to open the image picker. Images are grouped into **Inputs** and **Generations**. Select an image to add it to your message as a thumbnail chip such as `@input1` or `@generation2`. ## Write the instruction Add the instruction before or after the reference chips. For example, use one reference for a product and another for the model or setting where it should appear. When a message includes several references, identify the role of each image, such as the product, model, or setting. ## What's Next? - Use references while [creating or editing images](/using-fashn/agent/create-and-edit-images). - Choose a start image and an end image when [creating a video](/using-fashn/agent/create-videos#use-a-start-image-and-an-end-image). - [Upscale a referenced result](/using-fashn/agent/upscale-images). --- --- title: Transparent images description: Create images with transparency or remove the background of an existing image with Agent. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Transparent images Create images with transparency or remove the background of an existing image with Agent. Agent can create an image with transparency or remove the background of an existing image. Transparent results are PNG files. They appear in the conversation and are saved to Gallery. ## Create a transparent image Ask for transparency in your request. You can create the image using only text or from a product or model image, at up to 4K. > Put this dress on a model, on a transparent background. Transparency is not limited to the background. Parts of the subject or product can be semi-transparent too. Sheer fabric such as tulle stays see-through, so whatever you place behind the image shows through the fabric. Ask for transparency explicitly. A white, colored, or studio background is still a visible background. ## Remove a background Attach or [reference](/using-fashn/agent/reference-images) an existing image, then ask Agent to remove its background. > Remove the background. The result is the same size as the source image. Background removal cuts out the subject as it is, so it cannot make sheer fabric see-through. For that, create the image with transparency instead. ## Edit or upscale a transparent image Edits and [upscaling](/using-fashn/agent/upscale-images) keep the transparency. For example, “Make her T-shirt red” changes the T-shirt and leaves the transparent areas as they are. ## Limitations - Videos cannot be transparent. Transparent areas in a start image or an end image are filled with solid black or white. ## What's Next? - Open [Gallery](/using-fashn/gallery) to find the result, then [download or export it](/using-fashn/gallery/download-and-export) in a format that keeps transparency. - [Upscale the image](/using-fashn/agent/upscale-images) to increase its resolution. --- --- title: Upscale images description: Increase the resolution of an existing image with Agent. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Upscale images Increase the resolution of an existing image with Agent. Use upscaling to make an existing image larger while keeping its composition. Agent supports 2× and 4× upscaling and uses 2× by default. ## Upscale an image Attach or [reference](/using-fashn/agent/reference-images) the image you want to upscale, then ask Agent to upscale it. > Upscale this image. Request 4× explicitly when you do not want the default 2× option. > Upscale this image by 4×. When several images are available, identify the exact image you want to use. After a generation, a short follow-up can use the latest result when your choice is clear. The upscaled image appears in the conversation and is saved to Gallery. ## Resolution and limits | Option | What it does | | --- | --- | | 2× | Doubles the image's width and height. This is the default. | | 4× | Quadruples the image's width and height. You must request it explicitly. | The maximum output is 4K (~16MP). Agent cannot enlarge an image that is already at this limit. FASHN's image-resolution tiers map to total pixel counts: **1K (~1MP)**, **2K (~4MP)**, and **4K (~16MP)**. Exact dimensions depend on the image's aspect ratio. [Generation settings](/using-fashn/agent/generation-settings) control the resolution of a new image or edit. Upscaling enlarges an image that already exists. ## Recommended workflow Create and edit at 2K (~4MP), then upscale only your selected final results to 4K (~16MP). Working at 2K is faster and more cost-efficient, and generally makes complex edits easier to control, with results that follow your instructions more reliably. A result created directly at 4K can differ slightly in fine detail from a 2K result upscaled to 4K, but the gap is usually barely noticeable. ## What's Next? - [Create or edit an image](/using-fashn/agent/create-and-edit-images) before upscaling it. - [Create a video](/using-fashn/agent/create-videos) from your finished image. - Open [Gallery](/using-fashn/gallery) to find and download the result. --- --- title: Download and export description: Download original files or prepare images in WebP, JPEG, or PNG format. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Download and export Download original files or prepare images in WebP, JPEG, or PNG format. Images can be downloaded in their original format or exported as WebP, JPEG, or PNG. Videos download in their original format. ## Export images 1. To export one image, open it and select the download icon. To export several images, select them in Library or a Collection and select the download icon in the action bar. 2. Choose **Original**, **WebP**, **JPEG**, or **PNG**. For WebP and JPEG, adjust the quality when needed. 3. Review the image dimensions and estimated file size, then select **Export**. ## Choose export settings | Setting | Description | | --- | --- | | Format | **Original** keeps the source format. **WebP** creates compact web files, **JPEG** offers broad compatibility, and **PNG** is useful for lossless images or transparency. | | Quality | Available for WebP and JPEG. Higher quality increases file size; lower quality reduces file size but may introduce visible artifacts. | ## Export tips - For most web uses, start with WebP at 95% quality. Reduce the quality if you need a smaller file. - If the source image has transparency, choose Original, PNG, or WebP. JPEG removes transparency. - [Upscale the image](/using-fashn/agent/upscale-images) before exporting only when you need larger final dimensions. - If the exported image looks low-quality, increase the quality or choose Original. ## Download videos Open a video and select the download icon. To download several videos, select them and use the download action in the Gallery header. Videos keep their original format. ## What's Next? - [Upscale final images](/using-fashn/agent/upscale-images) before exporting when you need larger dimensions. - [Organize your Gallery](/using-fashn/gallery/organize) before preparing a final delivery. - Return to the [Gallery overview](/using-fashn/gallery). --- --- title: Gallery description: Use Agent, browse your Library, and find tools for organizing and exporting your work. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Gallery Use Agent, browse your Library, and find tools for organizing and exporting your work. The **Gallery** keeps the images and videos you create or import in FASHN. From here, you can work with Agent, browse your Library, and prepare your work for its next step. ## Open Gallery In the web app, select **Gallery** in the left sidebar. ## Browse your Library Library contains all images and videos in your current organization. New and imported items appear here automatically, with the newest items first. Use the page controls in the Gallery header to move between pages. To add images from your device, select **Import**. Import is available on paid plans, and imported images can be used like generated results. To group, classify, or filter your work, see [Organize your Gallery](/using-fashn/gallery/organize). ## Work with Agent in Gallery Use the [Agent](/using-fashn/agent) prompt at the bottom of Gallery to create images and videos or edit images without leaving the page. You can also drag an image from Gallery into the prompt to use it as a [reference](/using-fashn/agent/reference-images). The latest response appears above the prompt, and new results are added to Library. Open the response in Agent when you want to continue the full conversation. ## What's Next? --- --- title: Organize your Gallery description: Use Collections, Labels, Liked items, and filters to organize and find Gallery items. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Organize your Gallery Use Collections, Labels, Liked items, and filters to organize and find Gallery items. Use Collections, Labels, and Liked items for different stages of organizing and reviewing your Gallery. ## Choose an organization method | Option | Best for | | --- | --- | | **Collections** | Keeping assets from the same piece of work together, such as a product drop, marketplace submission, or client campaign | | **Labels** | Classifying items by shared visual traits or workflow states across Collections, such as front view, detail shot, approved, or needs retouching | | **Liked** | Marking promising results for a quick shortlist | ## Collections Collections keep selected images and videos together in named groups. An item remains in Library and can belong to more than one Collection. ### Create a Collection 1. Switch to the **Collections** tab. 2. Select **New Collection**. 3. Enter a name and confirm. ### Add items to a Collection In Library, select **Select**, choose one or more items, then select the Collection icon in the action bar. Choose an existing Collection or create a new one. On desktop, you can also right-click a single item and add it to a Collection. ### Browse and manage Collections Select a Collection to view its contents. You can use the same Gallery filters inside an open Collection. - **Rename**: Right-click a Collection and select Rename. - **Delete**: Right-click a Collection and select Delete. Deleting a Collection does not delete its items; they remain in Library. ## Labels Each item can have up to 10 Labels. ### Add Labels 1. In Library or an open Collection, select **Select**. 2. Choose one or more items. 3. Select the tag icon in the action bar. 4. Choose an existing Label, or select **Create label** to add a name and color. A new Label is applied to the selected items when it is created. ### Manage Labels Open the Labels menu from the selection action bar, then select **Manage labels**. From here, you can: - Rename a Label. - Change its color. - Delete it from the organization. Deleting a Label removes it from every item that uses it. It does not delete the items. ## Liked items Select the heart on a Gallery item to mark it as Liked. Select the heart again to remove the Like. To view your Liked items, open **Filters**, turn on **Liked**, and select **Apply**. ## Find items with filters Filters are available in Library and inside an open Collection. Select **Filters** to narrow the visible items by: - **Labels**: Show items with selected Labels. - **Liked**: Show items marked with the heart. - **Created by**: Show items created by selected organization members. This option appears when the organization has more than one member. - **Created with**: Show items created with selected tools or Agent. Choose your filters, then select **Apply**. To reset the view, select **Clear all**, then **Apply**. Selecting multiple Labels shows items with any selected Label. Different filter types work together, so adding **Liked** narrows those results to Liked items. ## Manage multiple items At the top of the page, choose **Select** to work with several items at once: - Select individual items or choose **Select All** - **Download**: Open [export options](/using-fashn/gallery/download-and-export) for the selected items - **Labels**: Add or remove Labels from the selected items - **Add to Collection**: Add selected items to a [Collection](#collections) - **Remove from Collection**: When viewing a Collection, remove selected items from it without deleting them - **Delete**: Permanently delete the selected items - Choose **Cancel** to exit ## What's Next? - [Download or export Gallery items](/using-fashn/gallery/download-and-export). - Learn how to [use Gallery images as references in Agent](/using-fashn/agent/reference-images). - Return to the [Gallery overview](/using-fashn/gallery). --- --- title: Guides description: In-depth guides covering model creation, virtual try-on, Agent workflows, product photos, prompting, model consistency, and parameter tuning in FASHN. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Guides In-depth guides covering model creation, virtual try-on, Agent workflows, product photos, prompting, model consistency, and parameter tuning in FASHN. Go beyond the basics with focused guides on getting the most out of FASHN. --- --- title: Background Remove description: Instantly isolate your subject with a clean, transparent or solid-color background. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Background Remove Instantly isolate your subject with a clean, transparent or solid-color background. Isolate your model or product for clean exports. **Background Remove** deletes the background from an image so you can export transparency or fill with a solid color. ## What is Background Remove? Background Remove deletes the background and leaves the subject intact. Choose a **transparent** background for overlays or a **solid color** to match your brand. ## When to Use Use Background Remove when you need a clean cutout of your subject for compositing, catalog listings, or brand-matched backgrounds. For replacing the background with a new scene, use [Edit](/using-fashn/studio/edit) instead. ## How to Use It ### Select Background Remove from the Toolbar Select **Background Remove** from the toolbar in Studio, then upload an image or choose one from your history. For best results, start with a clear, well-lit photo where the subject stands out from the background. ### Run Background Remove Click **Run**. The system detects the subject and removes the background in under 2 seconds. ### Choose Output After removal, choose how to export: - Keep a **transparent** background for ecommerce or compositing - Fill with a **solid color** to match your brand palette - Or continue in [Edit](/using-fashn/studio/edit) to set a new scene ### Prepare for Export or Continue Editing - Download or [export the image from Gallery](/using-fashn/gallery/download-and-export) - To keep the background transparent, export as PNG for full quality or WEBP for smaller file size with transparency preserved. - [Edit](/using-fashn/studio/edit) removes transparency, so make those edits first. [Upscale](/using-fashn/studio/upscale) keeps it. ## Parameters | Parameter | Description | | --------- | ------------------------------------------------------------------------------ | | Image | The image you want to process, uploaded or selected from history. | | Color | (When image is transparent) Choose a brand color or neutral tone for catalogs. | ## Limitations - High-quality background removal is optimized for images up to 2K resolution. - Fine edges such as hair strands or thin fabric fibers may be imperfectly detected, resulting in slight loss or background artifacts. - Sheer fabric such as tulle is not made semi-transparent. For that, [create a transparent image with Agent](/using-fashn/agent/transparent-images). - When an image contains multiple potential foreground elements (for example, a person and a table in front of them), the system may preserve all of them rather than removing secondary objects. ## Where Next? After removing the background, the post-result suggestion bar offers quick access to related tools: - [Export with transparency](/using-fashn/gallery/download-and-export) from Gallery - Continue refining with [Edit](/using-fashn/studio/edit) --- --- title: Create Model description: Build a reusable, brand-consistent AI model identity. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Create Model Build a reusable, brand-consistent AI model identity. Create a reusable AI model identity that stays consistent across visuals. Use Create Model to generate a model that matches your brand style and reuse it across Studio tools. ## What is Create Model? Create Model builds a reusable identity, not just a single image. Once generated, your model is saved to your library for use in future generations across campaigns and product lines. You can create a [Collection](/using-fashn/gallery/organize#collections) in your gallery named "Models" and save your creations in this folder for future usage. ## When to Use Use Create Model when you need a new model from scratch. If you already have a model photo and want to change the person, use [Model Swap](/using-fashn/studio/model-swap) instead. ## How to Use It ### Select Create Model from the Toolbar Select **Create Model** from the toolbar in Studio. ### Describe the Model Write a clear prompt that covers age, body type, expression, styling, and lighting. For example: "woman in her late twenties, medium skin tone, short wavy hair, calm studio pose under soft lighting." ### Add Optional References Add a **Face Reference** to guide identity or a **Pose Reference** to guide composition and pose. Then click **Run**. ### Refine or Continue After generation, the result opens in [Edit](/using-fashn/studio/edit) automatically. Use prompts to adjust pose, expression or background. ### Use in Other Tools Click **Use in Studio** to open [Try On](/using-fashn/studio/try-on) and apply products to this model. Use [Edit](/using-fashn/studio/edit) for background or framing changes. ### Export Download the image or [export it from Gallery](/using-fashn/gallery/download-and-export). ## Parameters | Parameter | Description | | ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Model prompt | Text that describes age, body type, expression, styling, and lighting. | | Face reference | Optional face photo to guide identity. See [Face Reference](/using-fashn/guides/model-consistency/face-reference) for plan details. | | Pose reference | Optional image to guide composition and pose. The AI will not copy the exact details from the image, but will use it as inspiration. | | Number of images | Choose 1-4 images to generate in one run. Outputs stay on-theme, are grouped as a single scrollable entry in Gallery, and credits scale with the number selected. | | Resolution | Auto (default), 1K, 2K, 4K. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | | Generation Mode | Auto (default), Fast, Balanced, Quality. Auto selects the optimal mode for your input. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | ## Best Practices - For full-body shots, include details about the hair and footwear in your prompt to help the AI generate a complete head-to-toe frame. - You can adjust the pose and background in [Edit](/using-fashn/studio/edit) after creating your model, so focus on perfecting the appearance first. ## Troubleshooting - **Cannot achieve full body shot (head cropped or feet not showing)**: Include details about the hair and footwear in your prompt to help the AI generate a complete head-to-toe frame. You can also use [Edit](/using-fashn/studio/edit) to zoom out with a prompt like "zoom out to show full body". ## Where Next? After creating your model, the post-result suggestion bar offers quick access to related tools: - Apply products with [Try On](/using-fashn/studio/try-on) or [Product to Model](/using-fashn/studio/product-to-model) - Adjust pose, background, or lighting with [Edit](/using-fashn/studio/edit) - Increase resolution with [Upscale](/using-fashn/studio/upscale) --- --- title: Edit description: Adjust pose, styling, background, and fixes while keeping identity and product fidelity. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Edit Adjust pose, styling, background, and fixes while keeping identity and product fidelity. Use **Edit** to refine generated images or new uploads. Adjust pose, swap backgrounds, add accessories or lighting, and fix small issues without losing the subject or product. Looking for a specific refinement? [Background Remove](/using-fashn/studio/background-remove), [Reframe](/using-fashn/studio/reframe), and [Face Swap](/using-fashn/studio/face-swap) are separate tools in the Studio toolbar. ## What is Edit? Edit is a post-processing tool that preserves identity and product fidelity while applying your instructions. It is ideal for: - Creating extra angles or subtle pose changes - Changing the background or mood of the image - Styling with accessories or different hairstyles - Fixing artifacts from Product to Model or Model Swap results ## When to Use Use Edit when you have an existing image and want to make targeted adjustments. If you need to change the entire background to a specific scene, see the [Background Change](#background-change) section below. ## How to Use It ### Select Edit from the Toolbar Select **Edit** from the toolbar in Studio, then upload an image or choose one from your history. ### Describe What to Change In the prompt box, state the edit in plain language. Examples: "turn the model slightly left," "add a black leather crossbody bag," or "soft daylight studio lighting." ### Add Optional Guidance - **Image Context:** Provide a reference image for pose, background, or texture when words are not enough. - **Annotate:** Scribble directly on the image to guide the edit (circle an area, draw an arrow, etc.). Your prompt should reference the annotation, for example: "remove the object I circled" or "change the color of the area I marked." ### Run Click **Run** to generate. You can generate 1-4 variations per run. ### Review and Iterate If something shifts too much, simplify the prompt or use annotations to target the change more precisely. ## Parameters | Parameter | Description | | ---------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Source image | The image you want to refine, uploaded or selected from history. | | Prompt | Freeform instructions for pose, styling, accessories, lighting, or fixes. | | Image Context | Optional reference image that guides pose, background, or texture when text alone is not enough. | | Number of images | Generate 1-4 outputs in one run. Credit use scales with the count. | | Resolution | Auto (default), 1K, 2K, 4K. Auto matches output to your input image resolution. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | | Generation Mode | Auto (default), Fast, Balanced, Quality. Auto selects the optimal mode for your input. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | ## Best Practices - Keep prompts targeted to one change at a time for predictable results. - For big changes, split the edit into two or more steps. For example, to go from a direct front view to a full side profile, first ask for a three-quarter view and then, from that result, ask for a side profile. - Use annotations to point the edit at a specific area instead of writing long, detailed instructions. - Add Image Context when you need to match a specific scene or angle that is hard to describe. ## Limitations - Edit keeps sizing stable, so changing hem or sleeve length or bag size usually will not take effect; address sizing at the source in Product to Model or Try On. ## Background Change Edit also handles full background replacement. Describe the scene you want and Edit will replace the background while preserving the model and product. **How to change the background:** 1. Select **Edit** from the toolbar and upload or choose your image. 2. In the prompt, describe the new background: "outdoor cafe in Paris with warm afternoon light," "clean white studio backdrop," or "tropical beach at golden hour." 3. Optionally add an **Image Context** reference showing the background style you want. 4. Click **Run**. **Tips for background changes:** - Keep the prompt focused on the background. Avoid mentioning the model or product unless you want those to change too. - Use Image Context for specific environments that are hard to describe in text. - For a simple solid or transparent background instead, use [Background Remove](/using-fashn/studio/background-remove). ## Where Next? After editing, the post-result suggestion bar offers quick access to related tools: - Increase resolution with [Upscale](/using-fashn/studio/upscale) - Adjust aspect ratio with [Reframe](/using-fashn/studio/reframe) - Remove or replace the background with [Background Remove](/using-fashn/studio/background-remove) - Swap the face with [Face Swap](/using-fashn/studio/face-swap) - Create motion clips with [Image to Video](/using-fashn/studio/image-to-video) --- --- title: Face Swap description: Update the model's identity while preserving outfit, pose, and lighting. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Face Swap Update the models identity while preserving outfit, pose, and lighting. Update the model's face while keeping outfit, body shape, pose, and lighting unchanged. Use Face Swap to localize, personalize, or match a creative direction without changing the product. ## What is Face Swap? Face Swap replaces the model's face in an existing image while keeping outfit, pose, lighting, and composition the same. Use it when the product image is already correct and only the face should change. Want to change the entire person to a different identity, while keeping the product, pose, and background the same? Use [Model Swap](/using-fashn/studio/model-swap) with Face Reference instead. ## When to Use Use Face Swap when the outfit, pose, and background are already correct and you only need to change the face. For changing the entire person, use [Model Swap](/using-fashn/studio/model-swap). ## How to Use It ### Select Face Swap from the Toolbar Select **Face Swap** from the toolbar in Studio, then upload an image or choose one from Gallery where the outfit and pose are correct. ### Add a Face Reference Choose a Portrait Element or add a custom face image. Use a reference with a similar angle and lighting. See the [Face Reference guide](/using-fashn/guides/model-consistency/face-reference) for reference options and plan access. ### Choose "Match Reference" or "Match Current Model" - "Match Reference" will keep the face reference's expression, head angle and gaze direction. This is recommended for maximum resemblance to the face reference. - "Match Current Model" will attempt to match the face reference to the current model's face (looking direction, expression). This is recommended for more natural results. Click **Run** to generate. ### Prepare for Export or Continue Editing Download, upscale, or [export the image from Gallery](/using-fashn/gallery/download-and-export). ## Parameters | Parameter | Description | | ------------------- | -------------------------------------------------------------------------------------------------------------------- | | Source image | The image you want to edit, uploaded or selected from Gallery. | | Face reference | Portrait Element or custom face photo used to control identity. Similar angle and lighting work best. | | Match Reference | Keep the face reference's expression, head angle and gaze direction. Recommended for maximum resemblance. | | Match Current Model | Adapt the face reference to the current model's face direction and expression. Recommended for more natural results. | ## Limitations Face Swap is currently limited to **2K resolution** output. ## Best Practices - Face Swap only changes the face, so the current model should already have a compatible visible skin tone and apparent age range for a natural transition across the face, neck, and other exposed skin. - If the visible skin tone or apparent age range differs substantially, use [Model Swap](/using-fashn/studio/model-swap) for a full-body identity change instead. - To increase the reliability of the swap, use a face reference with a similar angle and lighting. ## Troubleshooting If the result does not look quite right, try: - **Face size mismatch**: Use a face reference with a similar size to the subject. Match full-body images with full-body references and half-body images with half-body references. ## Where Next? After swapping the face, the post-result suggestion bar offers quick access to related tools: - Increase resolution with [Upscale](/using-fashn/studio/upscale) - Create more views or poses with [Edit](/using-fashn/studio/edit) - Adjust aspect ratio with [Reframe](/using-fashn/studio/reframe) - Create motion clips with [Image to Video](/using-fashn/studio/image-to-video) --- --- title: Image to Video description: Turn on-model images into short motion clips. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Image to Video Turn on-model images into short motion clips. Create short motion clips from finished on-model images. Use Image to Video to add subtle camera movement and garment motion for Reels, Shorts, or other video platforms. Studio requires a starting image. To create a video using only text, use [Agent](/using-fashn/agent). ## What is Image to Video? After you have produced a high-quality image (from Product to Model, Try-On, or Model Swap), Image to Video adds subtle camera moves, garment motion, and lighting shifts to produce a short clip. A starting image is required. ## When to Use Use Image to Video when you have a finished image and want to create a short video clip for social media or product pages. For best results, reframe your image to 9:16 first if targeting vertical video platforms. ## How to Use It ### Select Image to Video from the Toolbar Select **Image to Video** from the toolbar in Studio, then upload an image or choose one from your history. Pick a finished image where model, garment, pose, and background are set. ### Configure Settings Set your preferences: - **Duration**: 5 or 10 seconds - **Video Resolution**: 480p, 720p, or 1080p - **Prompt** (optional): short motion instruction - **End Image** (optional): reference image for the final frame See [Parameters](#parameters) for details. ### Generate Click **Run** to start generation. Processing time varies by image and server load. ### Review and Download When processing completes, the video appears in your history and in [Gallery](/using-fashn/gallery). If motion looks natural and product appearance is preserved, download via the controls. ### Publish Upload to your channels such as Reels, Shorts, or store banners. ## Parameters | Parameter | Description | | -------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Start image | Image to be used as the starting point for the video. Uploaded or selected from history. Higher resolution and clean backgrounds improve results. | | Duration | Length of the clip: 5 seconds or 10 seconds. | | Video Resolution | Output resolution and quality tier: 480p, 720p, or 1080p (Pro+). Each resolution uses a different AI model optimized for that fidelity level. See [Resolution](/using-fashn/guides/parameter-deep-dives/resolution#video-settings). | | Prompt (optional) | Short instruction to influence motion emphasis, for example "touch fabric" or "subtle camera pan". | | End Image (optional) | Second reference image that anchors the final frame of the video. See [End Image](#end-image). | ## End Image You can provide an optional **End Image** to anchor the final frame of the generated video. When an End Image is set, the clip smoothly transitions from the starting image toward the end image over its duration. ### Why use an End Image An End Image lets you specify transitions and movements that text prompts cannot reliably describe, such as a particular camera path, a specific final pose, or an exact scene composition. A common fashion use case is rotating a model to show the back of a garment. Without a reference, the video model has no way to know what the back actually looks like. With a back-view image as the End Image, the rotation renders the real garment instead of an invented one. ### How to set an End Image In the Image to Video tool, open the **End Image** optional input in the toolbar and upload or select a reference image. Then choose the duration and video resolution you want to use. ## Limitations - Videos cannot be transparent. Transparent areas in the start image or End Image are filled with solid black or white. ## Best Practices - Running without a prompt produces more natural, subtle motion. If you must add special instructions, keep them short and specific (for example, "raises arm to touch face"). - 5-second clips are easier to control and guide compared to 10-second clips. - Reframe your image to **9:16** with [Reframe](/using-fashn/studio/reframe) before generating video for Instagram Reels, TikTok, or YouTube Shorts. ## Troubleshooting If your video does not look as expected, try: - **Motion looks unnatural**: Pick a base image that allows for more natural movement. If you used a special instruction, simplify or remove it. Try a 5-second version first to check for improvement. ## Where Next? - [Create a video using only text with Agent](/using-fashn/agent/create-videos#create-a-video-using-only-text) - Download your finished video from [Gallery](/using-fashn/gallery) --- --- title: Studio description: All AI tools in one workspace. Generate, refine, and enhance images without switching between pages. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Studio All AI tools in one workspace. Generate, refine, and enhance images without switching between pages. The Studio brings all AI tools into one workspace. Select a tool from the toolbar, provide your inputs, and generate, all without leaving the page. Your history follows you across tools, so you can iterate and refine in one place. The Studio has two modes: the input screen where you set up a generation, and the result screen where you review and iterate. ### Before You Generate When you first open the Studio or click **Back** to start fresh: - **Upload Area** (center): The main screen shows options to Paste, Upload, or pick from your Gallery. You can also drag and drop an image directly. Some tools (like Product to Model) support uploading multiple images per generation. Click **Add another image** after the first upload to add more. - **Toolbar** (bottom): A horizontal bar with all available tools. Select the one you want before generating. - **Input Controls** (bottom): Below the toolbar, write an optional prompt, attach references (Face Reference, Image Prompt, Background), and set parameters like aspect ratio, resolution, generation mode, and number of outputs. ### Built-in Image Editor After uploading any image, hover over it and click the **Edit** icon (pencil) to open the built-in image editor. The editor has two tools: - **Crop**: Drag the handles to remove unwanted areas or reframe the subject. While in Crop mode, you can also rotate (in 90-degree increments or with the fine-adjustment slider), flip horizontally, and adjust the scale. - **Annotate**: Draw directly on the image to highlight or mark areas of interest. Click **Done** to apply your changes and return to the Studio. ### After You Generate Once a generation completes, the layout shifts to focus on your result and enable quick iteration: - **Result Viewer** (center): Your generated image at full size, with a download button in the corner. - **Input Summary** (left): Shows the references and inputs used for this generation. Click **Back** to return to the upload screen. - **History Panel** (right): Thumbnails of your recent generations across all tools. Click any thumbnail to load it. - **Suggestion Bar** (bottom): Replaces the toolbar with context-aware next steps like Edit, Image to Video, or Upscale. Click any suggestion and your result flows directly into that tool as input, no downloading or re-uploading needed. - **Iteration Prompt** (bottom): The prompt field changes to "What would you like to change?" so you can type a refinement and click **Run** to iterate without starting over. For example: generate with Product to Model, then click **Edit** to refine the background, then **Upscale** for high resolution, and finally **Image to Video** for motion. ## All Tools ### Generate Product to Model Generate on-model images from product photos. Try-On Virtually try products on specific model photos. Packshot Turn product or on-model images into clean packshots. Model Swap Replace the person while preserving outfit and pose. Create Model Build a reusable AI model identity from text. ### Refine Edit Adjust pose, styling, background, and fixes. Reframe Change aspect ratio by expanding or cropping. Background Remove Isolate subjects with transparent or solid backgrounds. Face Swap Change the face while keeping everything else. ### Enhance and Animate Upscale Increase resolution for sharper, professional results. Image to Video Turn images into short motion clips. ## Which Tool Should I Use? | Starting point | Goal | Tool | | ----------------------------------- | ------------------------------------- | ---------------------------------------------------------- | | Product photo (flat-lay, mannequin) | Put product on an AI model | [Product to Model](/using-fashn/studio/product-to-model) | | Model photo + product photo | Dress a specific model with a product | [Try-On](/using-fashn/studio/try-on) | | Product or on-model photo | Clean catalog packshot of the product | [Packshot](/using-fashn/studio/packshot) | | On-model photo | Change the person, keep the product | [Model Swap](/using-fashn/studio/model-swap) | | Any image | Adjust background, pose, or styling | [Edit](/using-fashn/studio/edit) | | Nothing (text only) | New model from scratch | [Create Model](/using-fashn/studio/create-model) | | Any image | Different aspect ratio | [Reframe](/using-fashn/studio/reframe) | | Any image | Clean or transparent background | [Background Remove](/using-fashn/studio/background-remove) | | Any image | Match a specific face | [Face Swap](/using-fashn/studio/face-swap) | | Any image | Higher resolution | [Upscale](/using-fashn/studio/upscale) | | Any image | Short motion video | [Image to Video](/using-fashn/studio/image-to-video) | --- --- title: Model Swap description: Swap models in existing images while preserving context. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Model Swap Swap models in existing images while preserving context. Swap the person while keeping the outfit, pose, lighting, and background the same. **Model Swap** changes the model’s appearance (identity, skin tone, hair, face) and preserves the product exactly as it appears. Use it when you want consistent product visuals with a different person representation. ## What is Model Swap? Model Swap is identity transformation. You are not changing the clothes or re-posing the subject. You replace or modify the person’s look while keeping everything else intact. It supports high resolutions and can match a consistent campaign model when needed. ### When to Use You have an existing image of a product worn by a person or mannequin and you want to replace the person while keeping the product unchanged. - Starting from a product photo without a person? Use [Product to Model](/using-fashn/studio/product-to-model). - Want to place a worn product onto another specific person's photo while keeping the pose and background? Use [Try-On](/using-fashn/studio/try-on). - Want a clean catalog packshot of the garment alone, without a model? Use [Packshot](/using-fashn/studio/packshot). ## How to Use It ### Choose Your Source Image Select an image where the outfit and pose are exactly how you want them. This image defines the outfit, pose, lighting, and background. ### Add Optional Guidance You can run with or without a prompt. To guide identity and styling, you can add: - Text Prompt to describe features like age, hair, skin tone, and general style - Face Reference to match a specific person (see [Face Reference guide](/using-fashn/guides/model-consistency/face-reference)) The product, pose, and background stay the same. You can change the background using a text prompt in this tool, but for higher quality results, use [Edit](/using-fashn/studio/edit) instead. It automatically adjusts lighting and shadows to match the new background. ### Run the Swap Click Run. Processing completes in seconds. ### Review and Adjust Review the result. Re-run for another option or refine your text prompt for identity details. ### Prepare for Export or Continue Editing Prepare your result for export or continue refining: - Download, upscale, or [export the image from Gallery](/using-fashn/gallery/download-and-export) - Adjust background or aspect ratio with [Edit](/using-fashn/studio/edit) and [Reframe](/using-fashn/studio/reframe) - Create a short video from this image with [Image to Video](/using-fashn/studio/image-to-video) ## Parameters Only the source image is required. Other inputs are optional and can be combined. | Parameter | Description | | ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Source image | An on-person or mannequin photo of your product with clear outfit, pose, and lighting. This defines the outfit, pose, and background to keep. | | Number of images | Generate 1–4 images in one run to compare multiple identity variations at once. | | Text prompt | Describe visible identity attributes such as apparent age, hair, skin tone, facial hair, and overall style. If no text prompt is provided, the system will make a subtle identity adjustment while keeping a similar apparent age range and visible skin tone. When used with a face reference, the text prompt can refine the appearance by adjusting details like hairstyle, expression, or gaze direction. | | Resolution | Auto (default), 1K, 2K, 4K. Auto matches output to your input image resolution. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | | Generation Mode | Auto (default), Fast, Balanced, Quality. Auto selects the optimal mode for your input. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | | Face reference | Face image to match identity. Useful for matching a specific person across a campaign. See [Face Reference](/using-fashn/guides/model-consistency/face-reference) for plan details. | ## Best Practices - Use a high-resolution (at least 1K), well-lit source image with minimal background clutter. - Focus on identity changes only; for pose or background changes, use [Edit](/using-fashn/studio/edit) instead. - For Face Reference, use a clear, front-facing photo with consistent lighting. See [Face Reference](/using-fashn/guides/model-consistency/face-reference) for reference options. ## Troubleshooting & Tips - **Incomplete body swap** → For major identity changes (for example, from light to dark skin tone), the system may miss updates on extremities such as hands or feet. If this occurs, include an additional instruction in the text prompt to update the skin tone in those areas. - **Partial mannequin completion** → When using a mannequin missing arms or lower body, the system may only swap visible sections. To complete the figure, add a text prompt instruction such as “add lower body with matching pants” or “add arms with matching skin tone.” ## Where Next? After swapping, the post-result suggestion bar offers quick access to related tools: - Increase resolution with [Upscale](/using-fashn/studio/upscale) - Create new poses or add branded scenes with [Edit](/using-fashn/studio/edit) - Fit marketplace or social aspect ratios with [Reframe](/using-fashn/studio/reframe) - Create motion clips with [Image to Video](/using-fashn/studio/image-to-video) --- --- title: Packshot description: Turn product or on-model images into clean, professional packshots. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Packshot Turn product or on-model images into clean, professional packshots. Create clean, e-commerce-ready packshots from almost any source photo. **Packshot** takes a product image (or even an on-model photo) and renders it as a polished catalog shot, automatically cleaning up wrinkles, twisted collars, and asymmetric drape. ## What is Packshot? Packshot turns everyday product photography into commercial-grade catalog imagery. With Packshot you can: - Start from a clean product, on-hanger, on-mannequin, or amateur on-floor photo - Start from an on-model photo and have the person automatically removed - Choose a presentation style (ghost mannequin, flat-lay, on-hanger, floating, and more) via the prompt or a reference image - Get crisp, pressed, symmetric results without manual retouching Use it for product listings, catalog hero shots, and marketplace assets. ## When to Use You have a product photo (or an on-model image) and you want a clean, catalog-ready packshot, not an on-model visual. - If you want the product worn by an AI model, use [Product to Model](/using-fashn/studio/product-to-model). - If you want to transfer the product onto a specific model photo, use [Try-On](/using-fashn/studio/try-on). - If you already have a packshot and just want small tweaks, use [Edit](/using-fashn/studio/edit). ## How to Use It ### Upload Your Source Image Upload the image you want to turn into a packshot. Packshot accepts a wide range of source styles, including clean product shots, on-hanger, on-mannequin, amateur on-floor photos, and on-model photos. When the source is on-model, the person is automatically removed and only the garment is rendered. ### Choose a Style and Refinements (Optional) The default presentation is ghost mannequin on a clean white background. To pick a different style, do one of the following: - **Write it in the prompt** (for example, "flat-lay packshot", "on-hanger", "floating garment", or "draped on a stone plinth"). - **Choose an Element from Clothing, Accessories, or Shoes** through [Image Context](#image-context) to use its presentation as a reference. You can combine both: a reference image sets the look, and the prompt refines details like "marble surface", "soft window-light shadows", or "top-down camera angle". Keep the prompt focused on staging and styling, not on changing the product itself. ### Generate Click **Run** to generate. You can produce 1-4 variations per run. ### Prepare for Export or Continue Editing Send your result straight into another tool, or download it from Gallery. - Download or [export the image from Gallery](/using-fashn/gallery/download-and-export) - Make small tweaks with [Edit](/using-fashn/studio/edit) - Increase resolution with [Upscale](/using-fashn/studio/upscale) - Fit marketplace ratios with [Reframe](/using-fashn/studio/reframe) ## Parameters Only the source image is required. Other inputs are optional and can be combined. | Parameter | Description | | ---------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Source image | The product or on-model photo to turn into a packshot. Supported sources include clean product, on-hanger, on-mannequin, amateur on-floor, and on-model photos. | | Text prompt | (Optional) Use the prompt to pick a presentation style ("flat-lay", "on-hanger", "floating") and refine staging ("marble surface", "window-light shadows", "top-down camera angle"). Keep the prompt focused on style, not on changing the product. | | Image Context | (Optional) Style reference for lighting, surface, composition, and mood. Upload an image, select one from Gallery, or choose an Element from Clothing, Accessories, or Shoes. | | Number of images | Generate 1-4 outputs in one run. Credit use scales with the count. | | Aspect ratio | Choose the output ratio. If you don't choose one, the result inherits the ratio from the reference image you supplied (Image Context; otherwise the source image). Supported: 1:1, 3:4, 4:3, 9:16, 16:9, 2:3, 3:2, 4:5, 5:4. | | Resolution | Auto (default), 1K, 2K, 4K. Auto matches output to your input image resolution. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | | Generation Mode | Auto (default), Fast, Balanced, Quality. Auto selects the optimal mode for your input and elevates to Balanced when an Image Context is provided. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | ## Packshot Styles Packshot does not use a preset selector. Style is driven by your prompt and your Image Context reference. The defaults match a clean ghost-mannequin packshot on a neutral background. Common styles you can request in the prompt: - **Ghost mannequin** (default): garment shaped by an invisible form with clean neck, sleeve, and hem openings. - **Flat-lay**: garment laid flat, viewed from above. Pairs well with surface and shadow descriptions like "on marble" or "with soft window-light shadows". - **On-hanger**: garment displayed on a visible hanger against a clean wall. - **Floating**: garment rendered as if suspended in air, with no visible support. You can also mix in named props or surfaces ("draped on a stone plinth", "on raw concrete floor", "with a hard editorial shadow") to steer composition and mood. ## Image Context Use **Image Context** when a style is easier to show than to describe, such as a specific surface, lighting, composition, or mood. Upload an image, select one from Gallery, or choose an [Element from Clothing, Accessories, or Shoes](/using-fashn/elements). ## Best Practices - Provide the cleanest source you have. Sharp focus and good lighting carry through to the packshot. - Keep the prompt focused on staging and styling (presentation, surface, lighting, camera angle). Avoid describing a different product. - Don't ask for cleanup explicitly. Wrinkles, twisted collars, asymmetric drape, and uneven hems are removed automatically. Construction, seams, prints, and proportions are preserved. - When you want a specific item from a multi-piece source, name it in the prompt (for example, "just the jacket", "only the shirt"). - For catalog detail pages or marketplace listings, generate at 2K or 4K so details stay sharp at high zoom. ## What's Supported - Source images: clean product, on-hanger, on-mannequin, amateur on-floor, and on-model photos - Presentation styles: ghost mannequin (default), flat-lay, on-hanger, floating, and freeform styling via the prompt - Item isolation: when the prompt names a specific item, only that item is rendered - All major wearable categories, including garments and accessories such as shoes, bags, hats, and eyewear - Vertical, square, and widescreen output formats ## Troubleshooting If your output doesn't look quite right: - **The wrong item was rendered** → Name the item explicitly in the prompt (for example, "only the jacket", "just the shoes"). - **The composition feels generic** → Add an Image Context reference to lock in the surface, lighting, or composition you want. - **The style isn't what you expected** → Name the presentation in the prompt ("flat-lay packshot", "on-hanger", "floating garment") instead of relying on the default. - **Wrinkles remain** → Cleanup is handled by default, but you can emphasize it in the prompt with instructions like "make the garment pressed, symmetric, and neatly arranged". ## Where Next? After generating, the post-result suggestion bar offers quick access to related tools: - Make small tweaks with [Edit](/using-fashn/studio/edit) - Increase resolution with [Upscale](/using-fashn/studio/upscale) - Fit marketplace or social aspect ratios with [Reframe](/using-fashn/studio/reframe) - Convert your new image into motion clips with [Image to Video](/using-fashn/studio/image-to-video) --- --- title: Product to Model description: Generate on-model images from product photos. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Product to Model Generate on-model images from product photos. Create on-model visuals from product photos. With **Product to Model**, upload a clothing or accessory photo and generate an image of that product on a model. ## What is Product to Model? Product to Model lets you: - Take a flat-lay, ghost mannequin, or hanger photo - Automatically apply that garment onto a generated model image - Generate on-model visuals suitable for ecommerce - Preserve details, texture, color, and fit Use it for ecommerce listings, social images, and lookbooks. ## When to Use You have a standalone, wearable product photo and you want to show the item on an AI fashion model. - If you need to place a product onto a specific existing model photo, use [Try-On](/using-fashn/studio/try-on). - If your source already shows a person or mannequin and you want to change the person while keeping the product unchanged, use [Model Swap](/using-fashn/studio/model-swap). - If you want a clean catalog packshot of the product instead of an on-model image, use [Packshot](/using-fashn/studio/packshot). ## Watch the Video Prefer video? This walkthrough covers everything from basics to advanced controls like Image Prompts, Background References, Face Reference, and resolution modes. ## How to Use It Start transforming your products into on-model visuals with these simple steps: ### Upload Your Product Images Upload a clear image of your product. You can upload up to three product images per generation to combine multiple items in a single look (for example, a top, bottom, and accessory). To add more products, click **Add another image** after uploading the first one. You can reorder or remove items before generating. After uploading, click the **Edit** icon on any image to crop, rotate, or refine it before generating. See [Built-in Image Editor](/using-fashn/studio#built-in-image-editor) for details. ### Add Optional Guidance Run with or without a prompt. To guide the result, you can add: - **Text Prompt** to describe the person, pose, styling, background, or lighting - **Image Prompt** to guide pose, environment, and lighting - **Background Reference** to set the scene - **Face Reference** to match identity See [Parameters](#parameters) for more details. ### Select Ratio & Resolution Select the ratio that fits your website layout, catalog, or social platform. Common options are 1:1 square, 2:3 or 3:4 for fashion catalogs. For resolution, pick the level that matches your use case. Bear in mind that the resolution you pick can affect generation credit cost. ### Generate the Image Click **Run** to generate your image. You’ll see your garment or accessory displayed on the model. ### Review and Adjust Examine the result. You can run the generation again for a different option or tweak your text prompt to switch up the pose or other composition specifics. ### Prepare for Export or Continue Editing Prepare your result for export or continue refining the image in [**Edit**](/using-fashn/studio/edit). - Download, upscale, or [export the image from Gallery](/using-fashn/gallery/download-and-export) - Adjust pose or background with [Edit](/using-fashn/studio/edit) - Create a short video from this image with [Image to Video](/using-fashn/studio/image-to-video) ## Parameters Only the product image is required. Other inputs are optional and can be combined. | Parameter | Description | | -------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Product image | Clear flat-lay, ghost mannequin, or on-hanger photo of a wearable product. Upload up to 3 product images per generation to combine multiple items in one look. | | Number of images | Generate 1–4 outputs in one run. Results follow the same prompt and appear as a single scrollable entry in Gallery. Credit use scales with the count. | | Text prompt | Describe model attributes, pose, background, lighting, and camera style. If you don't provide a text prompt, the system will create a fitting model and scene based on the product image. | | Image prompt | Visual reference for pose, environment, lighting, or photography style. Upload an image, select one from Gallery, or choose a Photography Style Element. | | Background reference | Backdrop for the result. Upload an image, select one from Gallery, or choose a Background Element. People in the reference are ignored. | | Face reference | Identity reference. Choose a Portrait Element or add a custom face image. See [Face Reference](/using-fashn/guides/model-consistency/face-reference) for reference options and plan access. | | Aspect ratio | Choose the output ratio. If you don’t choose one, the result inherits the ratio from the reference image you supplied (Background reference or Image prompt; otherwise Product image). Supported: 1:1, 3:4, 4:3, 9:16, 16:9, 2:3, 3:2, 4:5, 5:4. | | Resolution | Auto (default), 1K, 2K, 4K. Auto matches output to your input image resolution. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | | Generation Mode | Auto (default), Fast, Balanced, Quality. Auto selects the optimal mode for your input. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | ## Choose optional references from Elements The optional reference fields let you [browse Elements](/using-fashn/elements). Each field opens a category that matches its purpose: | Field | Opens in | | --- | --- | | Face Reference | Portraits | | Image Prompt | Photography Styles | | Background Reference | Backgrounds | Select another category when a different reference fits your goal better. ## Best Practices - Use clear images where the entire garment or accessory is visible, with even lighting and minimal background clutter. - Product photos should match or exceed the chosen resolution setting (for example, use 1K images for 1K output, 4K for 4K). - Provide product photos taken from a clean, top-down angle (flat lay style). Avoid tilted or uneven shots, such as garments photographed on the floor from a standing angle. - Avoid shadows, wrinkles, or cropped edges in product images. - Match the model’s pose to the garment type. Front-facing tops work best with front-facing models, for instance. The better your inputs, the more natural and photorealistic your results will be. ## What’s Supported - Upload up to 3 separate product images per generation. You can also include multiple products in a single image, but separate images give the best results for each item. - Works with flat-lay, ghost mannequin, and on-hanger product photos - Compatible with all major wearable categories, including garments and accessories such as hats, shoes, bags, belts, and eyewear - Supports vertical, square, and widescreen output formats ## Troubleshooting If your output doesn’t look quite right: - **Difficulty in generating full-body images** → Add details for footwear and hair appearance in the text prompt. Minor adjustments often improve results dramatically. ## Where Next? After generating, the post-result suggestion bar offers quick access to related tools: - Increase resolution with [Upscale](/using-fashn/studio/upscale) - Add branded or lifestyle scenes with [Edit](/using-fashn/studio/edit) - Fit marketplace or social aspect ratios with [Reframe](/using-fashn/studio/reframe) - Convert your new image into motion clips with [Image to Video](/using-fashn/studio/image-to-video) --- --- title: Reframe description: Adjust image aspect ratios by automatically expanding or cropping based on content. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Reframe Adjust image aspect ratios by automatically expanding or cropping based on content. Transform images to any aspect ratio while preserving important content. Reframe analyzes your image and automatically decides whether to expand (add new content) or crop (focus on key elements) to reach the target format. ## What is Reframe? Reframe automatically adjusts your image to a new aspect ratio by analyzing the content and choosing the best approach: - **Expansion**: When the target ratio requires more canvas, the AI generates new content that naturally extends backgrounds, surfaces, and textures. - **Cropping**: When the target ratio is more constrained, the AI identifies the most important subjects and trims sparse areas while keeping the visual focus intact. This makes it easy to prepare visuals for different platforms without manual editing or awkward cuts. ## When to Use Use Reframe when you need to adapt an image for a specific platform or layout, such as converting a portrait to landscape for a web banner, or a square to 9:16 for Reels and TikTok. ## How to Use It ### Select Reframe from the Toolbar Select **Reframe** from the toolbar in Studio, then upload an image or choose one from your history. ### Pick a Target Aspect Ratio Choose from the supported ratios: - 16:9, 5:4, 4:3, 3:2, 1:1, 2:3, 3:4, 4:5, 9:16 ### Generate Click **Run** to process the image. The AI will analyze your content and apply the optimal transformation. ### Save and Export Download the image or [export it from Gallery](/using-fashn/gallery/download-and-export). ## Parameters | Parameter | Description | | ------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Image | The image you want to reframe, uploaded or selected from history. | | Aspect ratio | Target ratio for the output. The AI determines whether to expand or crop based on your image content. Supported: 16:9, 5:4, 4:3, 3:2, 1:1, 2:3, 3:4, 4:5, 9:16. | ## Best Practices - Reframe to **9:16** before creating a video with [Image to Video](/using-fashn/studio/image-to-video) for Instagram Reels, TikTok, or YouTube Shorts. ## Troubleshooting If your reframed image does not look quite right, try the following: - **Subject duplicated**: When changing between extreme aspect ratios (for example, from 9:16 to 16:9), the system may duplicate the subject. If this happens, reframe in smaller steps: first to a square format, then to the wider aspect ratio. ## Where Next? After reframing, the post-result suggestion bar offers quick access to related tools: - Increase resolution with [Upscale](/using-fashn/studio/upscale) - [Download or export final files](/using-fashn/gallery/download-and-export) from Gallery - Create motion clips with [Image to Video](/using-fashn/studio/image-to-video) --- --- title: Try-On description: Virtually try products on models with consistent results. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Try-On Virtually try products on models with consistent results. Transfer clothing and other wearable items from one photo to another. Try-On preserves product details, texture, and fit while matching the target model's pose and body shape. ## What is Try-On? Virtual try-on shows how a specific item looks on a specific person. Unlike Product to Model, which generates a new AI fashion model to showcase the product, Try-On keeps your chosen model, pose, and body shape while swapping in the garment or accessory. ### When to Use Use Try-On when you want to place a product on an existing model photo while keeping that person's identity, pose, and background. It works with tops, bottoms, dresses, outerwear, shoes, hats, scarves, and other wearable fashion items. - Want a new AI-generated model rather than using an existing model image? Use [Product to Model](/using-fashn/studio/product-to-model). - Already have the product on a person and only want to change who it is? Use [Model Swap](/using-fashn/studio/model-swap). - Want a clean catalog packshot of the garment alone, without the model? Use [Packshot](/using-fashn/studio/packshot). ## How to Use It ### Choose or Upload a Model Image Upload a photo of the person who should wear the item, select one from Gallery, or choose a Model Element. ### Upload the Product Image Upload the product you want to try on. On-model, flat-lay, and ghost mannequin photos are all supported. ### Adjust Settings (Optional) - **Prompt:** add instructions to guide the generation, such as "open jacket", "tucked shirt", or other styling details. If the product image contains multiple items (for example, a top and bottom) and you want to transfer only one, specify the item in the prompt. - **Number of Images (1-4):** generate multiple variations in a single run. ### Run the Try-On Click **Run** to start. Once the generation completes, browse through your results using the navigation arrows or thumbnails. ### Prepare for Export or Continue Editing - Download, upscale, or [export the image from Gallery](/using-fashn/gallery/download-and-export) - Fix mistakes, adjust background, or change aspect ratio with [Edit](/using-fashn/studio/edit) and [Reframe](/using-fashn/studio/reframe) - Create a short video from this image with [Image to Video](/using-fashn/studio/image-to-video) ## Parameters Only the model image and product image are required. Other inputs are optional. | Parameter | Description | | ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Model image | The person who will wear the item. Upload your own photo, select one from Gallery, or choose a Model Element. See [Best Practices for Model Image](#best-practices-for-model-image). | | Product image | The wearable item to apply. On-model, flat-lay, and ghost mannequin photos are supported. See [Best Practices for Product Images](#best-practices-for-product-images). | | Prompt | (Optional) Add instructions to guide the generation. For example: "open jacket", "tucked shirt", "rolled up sleeves". | | Resolution | Auto (default), 1K, 2K, 4K. Auto matches output to your input image resolution. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | | Generation Mode | Auto (default), Fast, Balanced, Quality. Auto selects the optimal mode for your input. See [Resolution & Generation Mode](/using-fashn/guides/parameter-deep-dives/resolution). | | Number of Images | (1-4) Generate multiple variations in a single run. | ## Choose a model from Elements Select **Elements** in the model image field and choose **Models** to try the product on a curated person. See [Elements](/using-fashn/elements) for the available reference types and access details. ## Best Practices for Model Image **Recommended:** - Use a 2:3 aspect ratio for balanced framing. - Ensure the full outfit is visible and not cropped. - Choose fitted clothing so the system can detect edges and body shape. - Use three-quarter or full-body shots for best alignment. - Keep the pose natural, such as standing or slightly angled. **Avoid:** - Extreme poses or unusual perspectives. - Photos with multiple people or children. - Hair or objects covering the clothes. - Wide aspect ratios or distant shots. ## Best Practices for Product Images **Recommended:** - On-model photos deliver the best results, as garments appear in their natural shape and lighting. - Ghost mannequin shots are a good alternative when model photos are unavailable, as they preserve the garment's shape and proportions. **Acceptable but less optimal:** - Flat-lay images can work, but the lack of depth may reduce realism, especially for structured items like shoes or bags. **Avoid:** - Apparel on hangers: introduces distortion and loses shape context. - Apparel on the floor: wrinkles and uneven draping reduce output quality. ## Troubleshooting - **Original clothing is not being removed correctly** → add a specific instruction to remove the unwanted item (e.g., "remove the jacket"). - **Product details are not transferring accurately** → use a higher quality source image with clear outlines and good lighting. On-model photos typically produce the best results. - **Results don't match expected styling** → add instructions in the prompt field (e.g., "tucked shirt", "open jacket"). ## Where Next? After generating, the post-result suggestion bar offers quick access to related tools: - Increase resolution with [Upscale](/using-fashn/studio/upscale) - Fit marketplace or social aspect ratios with [Reframe](/using-fashn/studio/reframe) - Create motion clips with [Image to Video](/using-fashn/studio/image-to-video) --- --- title: Upscale description: Increase resolution for sharper, professional results. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Upscale Increase resolution for sharper, professional results. Increase image resolution by **2x** or **4x**, or use **Enhance** to improve quality without changing size. ## What is Upscale? Upscale improves the resolution and clarity of generated images. Use it to prepare visuals for storefronts, social feeds, or large formats. You can choose between **2x**, **4x**, and **Enhance** options, with a maximum output resolution of **4K**. ## When to Use Use Upscale as the last step before exporting. Many tools render at 1K, and upscaling ensures your final images are sharp enough for their intended use. ## How to Use It ### Select Upscale from the Toolbar Select **Upscale** from the toolbar in Studio, then upload an image or choose one from your history. ### Choose an Upscale Option Select **2x**, **4x**, or **Enhance** and click **Run**. Processing starts automatically. - **2x**: Doubles resolution for sharper web use. Enhances detail such as skin texture, iris, hair, and fabric while keeping results realistic. (Recommended for most use cases.) - **4x**: Produces high-resolution output for full-screen or print (up to 4K). May add imagined details but can reveal more texture and depth when it works well. - **Enhance**: Improves image quality by adding detail, refining textures, and sharpening edges without changing the resolution. Useful for images already at 4K where further upscaling is not possible. ### Review and Save When processing completes, the upscaled version appears in your history and in [Gallery](/using-fashn/gallery). Download it in original quality or [export it in another format](/using-fashn/gallery/download-and-export). ## Parameters | Parameter | Description | | ---------------- | -------------------------------------------------------- | | Image | The image to upscale, uploaded or selected from history. | | Upscaling factor | Choose **2x**, **4x**, or **Enhance**. | ## Limitations - Up to 4K resolution is supported. ## Best Practices - 2x upscale is more reliable for most images; 4x can add more detail but may introduce artifacts. - Use Enhance when your image is already at 4K or has sufficient resolution but needs higher visual quality. - Treat upscaling as the last step before export. Many tools render at 1K. If you open an upscaled image in [Edit](/using-fashn/studio/edit) again, it will be processed at 1K and you will need to upscale again before exporting. - For web publishing, [export from Gallery](/using-fashn/gallery/download-and-export) as WebP at 95% quality. ## Troubleshooting If your upscale does not look right, try the following: - **Distortion or artifacts**: If using 4x upscale, switch to 2x. If already on 2x, check whether the distortions were present in the original image and were simply amplified during upscaling. ## Where Next? After upscaling, the post-result suggestion bar offers quick access to related tools: - [Download or export final files](/using-fashn/gallery/download-and-export) from Gallery --- --- title: Agent Workflows description: Go from product photos to finished images and video in one Agent conversation. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Agent Workflows Go from product photos to finished images and video in one Agent conversation. A workflow is a chain of short messages in one Agent conversation. Follow along with your own product, from the first photo to the finished photo shoot. --- --- title: Mannequin to Photo Shoot description: Turn two mannequin photos into a photo shoot in one Agent conversation, with front and back views, a close up, a new color, and a video, using the exact messages from the video. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Mannequin to Photo Shoot Turn two mannequin photos into a photo shoot in one Agent conversation, with front and back views, a close up, a new color, and a video, using the exact messages from the video. This guide shows you how to turn two mannequin photos into a photo shoot in **one conversation** with [Agent](/using-fashn/agent): a front and back view, a close up, a new color, and a video. Every step uses the exact message from the video, so you can follow along with your own product. Every example below was made in Agent at **4:5**, **2K**, **Quality** (set in [Generation settings](/using-fashn/agent/generation-settings)). ## What you need The first message uses four images: the front and back of **your product**, a portrait, and shoes. Mannequin photos are used here, and [on-person and flat-lay photos](/using-fashn/guides/virtual-try-on/getting-started-with-virtual-try-on#product-photo) work too. Your product (front)Your product (back)PortraitShoes You can add your own portrait and shoes, or pick them from [Elements](/using-fashn/elements). To use Elements, open **Add images** in Agent, select **Browse Elements**, and choose from the **Portraits** and **Shoes** tabs. ## Step 1: Front and back Attach all four images and ask for both views in one message. New conversation > Put this dress and these shoes on the model. Create a front and back view. FrontBack Always give Agent the front and the back of your product, so it does not have to guess what the back looks like. ## Step 2: Close up For a detail shot, stay in the same conversation and ask for a close up. Continue the conversation > Make a close-up of the embroidery on the front. ## Step 3: New color Now change the color and the embroidery in all three images with one message. Continue the conversation > Make the dress ivory with gold flowers with no leaves in all three images. FrontBackClose up ## Step 4: Video Last, turn the close up into a short video for your product page. Continue the conversation > Make a 5 second video of the close-up with camera movement. ## What to remember - **Front and back:** give Agent both product photos so it does not have to guess. - **One conversation:** stay in the same conversation, so each message builds on the last result. - **Short messages:** say what you want in plain words, and Agent writes the detailed prompt. - **All at once:** one message can change every image in the conversation. ## Keep going - Need your own model first? Read [How to Create a Realistic AI Model](/using-fashn/guides/model-creation/how-to-create-a-realistic-ai-model). - Want to dress a model image you already have? Read [Getting Started with Virtual Try-On](/using-fashn/guides/virtual-try-on/getting-started-with-virtual-try-on). - For more on Agent, see [Create and edit images](/using-fashn/agent/create-and-edit-images) and [Create videos](/using-fashn/agent/create-videos). --- --- title: Face Reference description: Keep the same model identity across Product to Model, Model Swap, Create Model, and Face Swap. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Face Reference Keep the same model identity across Product to Model, Model Swap, Create Model, and Face Swap. Face Reference is a dedicated feature that maintains identity across **Product to Model**, **Model Swap**, **Create Model**, and **Face Swap**. It keeps generations aligned with a single face image and works more reliably and easily than methods like fine-tuning or training on a set of images. Every paid App plan includes Face Reference. Use a Portrait Element, upload a face photo, or choose one from Gallery. ## Why Face Reference instead of fine-tuning Earlier versions of FASHN supported consistency through fine-tuning (custom training) on about 8 to 12 images of an individual. Although training can work, it is difficult to do well. Each source image can introduce bias, and the underlying model becomes less flexible after fine-tuning. Customers also found it challenging to gather a complete and consistent training set in the first place, since they needed consistent results before they could produce the required images. After several months of testing with customers, we found that anchoring identity to a single face reference produces more reliable results, avoids dataset collection issues, and maintains the flexibility of the base model. ## How Face Reference works Face Reference influences supported tools in two ways: - It guides the base generation to match visible appearance traits from the reference, including skin tone and complexion, facial features, hair, and apparent age range. - It reinforces the likeness with special focus on facial features. Face Reference adapts identity to new angles, expressions, hairstyles, and lighting. It keeps the identity consistent while still allowing the model to generate natural, flexible, and creative variations. ## Where to use it - **[Product to Model](/using-fashn/studio/product-to-model)**: Generates a model that reflects the face reference across the full body. Visible traits such as skin tone, complexion, and apparent age range remain consistent across the face and body. - **[Model Swap](/using-fashn/studio/model-swap)**: Replaces the existing model with the face reference and updates exposed skin and other visible appearance traits across the body to remain consistent with it. - **[Create Model](/using-fashn/studio/create-model)**: Builds a new model from scratch that resembles the face reference, with full-body alignment from the start. - **[Face Swap](/using-fashn/studio/face-swap)**: Includes a Face Swap variant focused on restoring identity rather than altering it. Use it when pose, background, or other edits distort the face. It can also swap identities, but the underlying model must already match the new face in skin tone and age. ## Additional controls Face Reference adds a control that adjusts how strongly the result follows the reference image versus the base image or prompt: - All supported tools show **Match Reference**. - You will also see **Match Base** or **Match Prompt**, depending on whether the input starts from an image or a prompt. **Example (Model Swap):** If the current model faces left and smiles, but the face reference looks straight ahead with a neutral expression, the control lets you decide whether to keep the existing pose and expression or follow the reference more closely. **Example (Product to Model):** The control decides how much the prompt affects the result. For instance, a prompt such as "a model looking up" can shift pose and expression unless you set the control to follow the reference image more strongly. ## Preparing reference photos - Even though the feature can adapt to new angles or expressions, results improve when the reference photo already resembles the expected output. In these cases, keep the **Match Reference** control at its default setting to maximize resemblance. - Hair, makeup, and accessories should be close to the desired final look. - Avoid heavy occlusions such as sunglasses, masks, or hands covering the face. ### Choose a reference source - **[Portrait Elements](/using-fashn/elements#choose-the-right-type-of-element)**: Choose a curated portrait directly from the Face Reference field. - **Custom reference images**: Upload a face image or select one from Gallery. If needed, [upscale a custom reference](/using-fashn/agent/upscale-images) before using it to improve facial detail. ## Effect on runtime and cost Face Reference is computationally heavy and increases both generation time and credit cost. In the Studio, enabling Face Reference adds **+2 credits** per output image. See [App Pricing](/plans-and-pricing/app-pricing#credit-costs) for the full cost breakdown. ## Limitations Face Reference works best at **2K resolution or below**, where identity and product details are retained most reliably. You can still select 4K, but fine facial and product features may be harder to preserve at higher resolutions. ## Workflow summary Choose a Portrait Element, upload a face image, or select one from Gallery. Use the tool that matches your goal. **Model Swap** updates existing model photos, while **Product to Model** generates new on-model images from scratch. Use **[Face Swap](/using-fashn/studio/face-swap)** to restore resemblance if identity drifts after pose or background changes. --- --- title: How to Keep Your Models Consistent description: Maintain the same model appearance across sessions using Product to Model and supporting tools. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # How to Keep Your Models Consistent Maintain the same model appearance across sessions using Product to Model and supporting tools. ## Introduction Maintaining the same model appearance is important for many brands. It helps create a familiar visual style across PDPs and campaigns, similar to how real photoshoots rely on consistent talent and controlled conditions. The FASHN app offers several ways to achieve model consistency. This article focuses on the **[Product to Model](/using-fashn/studio/product-to-model)** tool, since it generates a new model each time you apply it to a product. We will look at how to achieve consistency across different sessions, and why some situations are easier than others. There are two levels of difficulty when targeting model consistency: 1. Consistency using the **same product** 2. Consistency across **different products** The first is easier. The second is more challenging because product images influence the AI and can shift the appearance of the generated model. --- ## Consistency With the Same Product ### Use detailed prompts and generate multiple images at once Start by writing a detailed prompt that describes the model. Focus on physical appearance. Keep backgrounds simple (for example: white) to avoid unnecessary variation. Generate four images at once using **[Product to Model](/using-fashn/studio/product-to-model)**. The four outputs may not match perfectly, but a strong model description usually helps the system converge on a specific look, creating four variations of the same model in slightly different poses or compositions. These outputs become the base set for further expansion. ### Use Edit to change poses The [Edit](/using-fashn/studio/edit) tool, especially at **Precise 1K** resolution, can adjust poses while keeping both model appearance and product details consistent. Changes should be moderate and applied in steps. For example, move from a frontal view to a three quarter view before attempting a full side view. You can direct the shoot with prompts such as: - hand resting on hip - relaxed arms at sides - hand touching face, smiling This makes it possible to build a full range of poses similar to a real photoshoot. ### Handling unique side or back details Side or back variations are more difficult. For small details such as slits or vents, you can prompt them directly, for example: “three quarter view, side slit visible”. If the back is significantly different (for example, a hoodie with a graphic print), create new Product to Model images using the same prompt but specifying a back view. Since the face is not visible, a consistent body description and hairstyle can make the identity appear continuous. --- ## Consistency Across Different Products ### Why this is harder The final model appearance comes from a combination of the product image and your prompt. Changing the product changes the visual cues the AI receives, so even identical prompts may produce different model identities. This is especially noticeable with category differences, such as womenswear vs menswear or swimwear vs outerwear. To work around this, you need a method that preserves a specific identity regardless of the product. ### Using Try On Use **Try On** to keep the exact person, pose, and background while swapping in a new product. It’s the fastest way to connect an existing model photo with a different garment or accessory without changing their identity. After you get the result, move it to [Edit](/using-fashn/studio/edit) if you want to adjust the pose or background for variety. Keep these points in mind: - Expect to retry more often than usual. - If a starting model photo fails repeatedly, switch to a simpler pose or clearer framing. The priority is to combine the model identity with the product correctly. - You can use full body, upper body or face photos as starting points. The goal is to achieve at least one successful transfer of the product onto an existing model image, either from past campaigns or earlier generations. Once you have that anchor image, you can return to the **Edit** tool for pose and background changes, following the same steps used for single product consistency. --- ## Plan Upgrades That Can Help ### Model and Resolution Variants (Pro) The 4K versions of **Product to Model** are better at handling try on operations. **Precise 1K** can produce very accurate results but may fail to perform the transfer and simply return your original model image. **Balanced 4K** and **Creative 4K** are more likely to perform the try on, although success is not guaranteed on the first try. ### Face Reference **Face Reference** is our dedicated identity consistency feature. It works across **Product to Model**, **Model Swap**, **Create Model**, and **Face Swap**, and is the most reliable and straightforward way to maintain the same model across many different products. See the [Face Reference guide](/using-fashn/guides/model-consistency/face-reference) for reference options and plan access. --- Continue to the next article to learn more about [Face Reference](/using-fashn/guides/model-consistency/face-reference). --- --- title: Model Consistency description: Keep AI-generated models looking like the same person across products and sessions. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Model Consistency Keep AI-generated models looking like the same person across products and sessions. Read our guides to learn how to keep your models' identities consistent across products and sessions. ## Further Reading For a step-by-step walkthrough with screenshots, see our blog post: [How to Create Consistent On-Model Product Images with AI](https://fashn.ai/blog/how-to-create-consistent-on-model-product-images-with-ai) --- --- title: How to Create a Realistic AI Model description: Three ways to create a realistic AI fashion model in Agent, with text, from Elements, or from a reference image, using the exact prompts from the video. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # How to Create a Realistic AI Model Three ways to create a realistic AI fashion model in Agent, with text, from Elements, or from a reference image, using the exact prompts from the video. This guide shows you three ways to create a realistic AI fashion model: by describing the model in **text**, by starting from a model in [**Elements**](/using-fashn/elements), or by starting from a **photo of a real person**. Every step uses the exact prompt from the video, so you can follow along in [Agent](/using-fashn/agent). By the end you will have a model you can dress, reshoot, and reuse across your products. Every example below was made in Agent at **2K**, **Balanced** (set in [Generation settings](/using-fashn/agent/generation-settings)). ## Method 1: Text Use text when you have no one in mind yet. You start with a short prompt, look at what comes back, and steer from there, one detail per step. ### Start with the default You do not need a long prompt. Three words give you a clean, realistic model on a white background, ready to work with. New conversation > Create a model ### Start with one detail Same three words, plus one detail. A new model comes back with red hair. New conversation > Create a model with red hair ### Describe the model (5 categories) Be specific. These five details do most of the work: - Gender - Ethnicity - Age - Hair - Body New conversation > Create a model. She is an East Asian woman in her early twenties with a bob hairstyle and a slim build. ### Change the details Stay in the same conversation and tell Agent what to change. The model stays the same person while the clothes and hair change. Continue the conversation > She should be wearing a red tracksuit and running shoes. Her hair should be long and slicked back. ### Change the camera angle Still in the same conversation, ask for a different camera angle. The model and clothes stay the same, only the camera moves. Continue the conversation > Close up portrait of her face ## Method 2: Elements Library When you want ideas, [Elements](/using-fashn/elements) is a good place to start. Open **Add images** in Agent, select **Browse Elements**, and pick a model. She's attached, so this prompt edits her. She stays the same person and only the clothes change. New conversation > She should be wearing a white turtle neck, red patent leather mini skirt, and white leather boots. ElementsResult ## Method 3: Reference Image When the model has to be a specific person, start from a photo. A sharp, well lit photo with the face clearly visible gives you the best result. New conversation > Make him a fashion model, full body. ReferenceResult ### Restyle from the original photo To change the outfit, reference the original photo instead of the image Agent just made. Each new generation built on an old one loses some quality. In the media panel, select the [**@** button](/using-fashn/agent/reference-images) on your photo and describe the change. Continue the conversation > @input1 Make him a fashion model, full body, relaxed pose. He should be wearing a cream knit t-shirt, brown pleated trousers, and black boots. ### Close up For a close up, stay in the same conversation and describe the edit. Continue the conversation > Close up editorial side shot, he is looking at the camera with an intense gaze ## Which method to use - **Start from scratch:** describe the model with text. - **Need some ideas:** start from [Elements](/using-fashn/elements). - **Need this exact person:** start from a reference image. ## Keep going - To reuse the same model across many products, read [How to Keep Your Models Consistent](/using-fashn/guides/model-consistency/how-to-keep-your-models-consistent) and [Face Reference](/using-fashn/guides/model-consistency/face-reference). - For more on writing prompts, see [Prompting in FASHN](/using-fashn/guides/prompting-in-fashn). --- --- title: Model Creation description: Design the exact AI model you want, then keep using it across your products. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Model Creation Design the exact AI model you want, then keep using it across your products. Casting directors look through hundreds of models to find the right one. In FASHN you design the model you want, then reuse them across your catalog. --- --- title: Parameter Deep Dives description: Understand cross-cutting parameters like resolution and how they affect outputs. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Parameter Deep Dives Understand cross-cutting parameters like resolution and how they affect outputs. In-depth guidance for parameters that appear across multiple tools, so you can pick the right setting for quality, speed, and cost. --- --- title: Resolution & Generation Mode description: This page explains how resolution and generation mode work in the FASHN app for both images and videos. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Resolution & Generation Mode This page explains how resolution and generation mode work in the FASHN app for both images and videos. This page explains how **resolution** and **generation mode** work in the FASHN app. These are two independent controls that together determine the output quality, speed, and credit cost of your generations. If you are unsure what to choose, leave both on auto and the app will select the best settings for your input. ## Image Settings ### Resolution Resolution controls the output pixel count of your generated images: - **Auto** (default) — matches output resolution to your input image - **1K** (~1 megapixel) - **2K** (~4 megapixels) - **4K** (~16 megapixels) Exact output dimensions depend on the resolution tier and the image aspect ratio. All plans have access to all tiers. **Auto** compares the total pixel count of your input image to the three tiers and picks the closest match. Tools without an image input, such as Create Model, default to 1K. **1K** is the fastest and most affordable option. It covers more use cases than you might expect: social media platforms like Instagram, TikTok, and Pinterest display images at around 1,080px wide, and most e-commerce platforms serve product listing images at 300-800px. If your output is headed for social media or product grids, 1K is all you need. **2K** is the sweet spot for e-commerce product detail pages. Major platforms like Shopify, Amazon, and Zalando recommend uploading images of at least 2,000px on the longest side to support product zoom, and 2K lands right in that range. Use 2K when customers need to inspect product textures, stitching, or fabric detail through a zoom or magnifier feature. **4K** produces the highest detail output. Most web and social channels will never display this resolution, so 4K is best reserved for print catalogs (A4 at 300 DPI needs roughly 9 megapixels), large-format signage, or when you need a single high-resolution master to crop into multiple aspect ratios and deliverables. ### Generation Mode Generation mode sets the quality level for image generation: - **Auto** (default) — selects the optimal mode for your input - **Fast** — fastest generation, most affordable - **Balanced** — middle ground between speed and quality - **Quality** — most capable model, most realistic results (Pro and Agency plans only) **Auto** chooses between Fast and Balanced based on factors like the requested resolution and the tool, leaning toward Balanced when the context benefits (higher resolutions or multi-image tools like Try-On) and Fast otherwise. Quality is opt-in only and must be selected manually when you want the highest fidelity (Pro and Agency plans only). **Fast** is the quickest option and produces consistent, predictable output. It follows instructions precisely and is the most reliable choice for localized edits (adjusting hand position, tilting the head, small fixes) without altering the rest of the image. **Balanced** delivers sharper detail and richer textures while remaining responsive. It handles multi-image inputs (Try-On, Edit with Image Context) better than Fast and follows detailed instructions more reliably. A good default for most production workflows. **Quality** produces the most photorealistic output with the highest level of detail. It excels at maintaining subject consistency when working with multiple images (Try-On, Edit with Image Context) and interprets nuanced prompts with the highest fidelity. Takes longer to generate but delivers the best results for final, publication-ready assets. ### Image credit costs Credit costs depend on the combination of resolution and generation mode. If you request multiple outputs in a single run, the cost is multiplied by the number of outputs. These costs apply to Product to Model, Model Swap, Create Model, Edit, Reframe, Try-On, and Packshot: | Generation Mode | 1K | 2K | 4K | | --------------- | --- | --- | --- | | Fast | 1 | 1 | 1 | | Balanced | 1 | 1 | 2 | | Quality | 1 | 2 | 3 | **Example:** Model Swap at 2K Quality, requesting 2 images: 2 x 2 = 4 credits. ## Video Settings Video resolution works differently from image resolution. Instead of 1K/2K/4K tiers, video uses standard video formats: - **480p** - **720p** - **1080p** Each resolution corresponds to a different AI video model optimized for that fidelity level. Higher video resolution also implies better quality: more realistic motion, better prompt adherence, and more consistent subjects across frames. ### Video credit costs | Duration \ Resolution | 480p | 720p | 1080p | | --------------------- | ---- | ---- | ----- | | 5 seconds | 1 | 3 | 6 | | 10 seconds | 2 | 6 | 12 | ### Plan-based video resolution access - **Basic plan**: up to 720p - **Pro and Agency plans**: up to 1080p --- --- title: How to Photograph Clothes for AI description: Four things that make your product photos work in Agent: views, fit, light, and details. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # How to Photograph Clothes for AI Four things that make your product photos work in Agent: views, fit, light, and details. This guide shows you how to photograph your clothes so [Agent](/using-fashn/agent) gets them right. Almost any photo works, and four things make it better: **views**, **fit**, **light**, and **details**. Every example comes from one jacket shot three ways under four kinds of light. Every example below was made in Agent at **2:3**, **2K**, **Balanced** (set in [Generation settings](/using-fashn/agent/generation-settings)). Each one used the same message: "Put this jacket on him. Create three separate images: a front, a side and a back view." ## Views Shoot the **front and the back** of your product. If you only give Agent the front, it has to guess what the back looks like. With only the front photo, the back came out wrong: Your photo (front only)Back view from Agent With a back photo added, the back came out right. A side photo is a bonus that shows details like the stripes down the sleeves. FrontSideBackFrontSideBack ## Fit How the product is photographed changes how well Agent understands its fit. 1. On a person2. Mannequin3. Flat lay 1. **On a person** is best, because it shows how the piece sits on a body. 2. **A mannequin** is second best. It gives the shape of a body, but it is stiffer than a real person. 3. **A flat lay** still works, but Agent has to guess the fit. ## Light The jacket was photographed under four kinds of light: warm bulbs, neutral bulbs, daylight, and the flash on a phone. WarmNeutralDaylightFlash **Neutral light** and **the flash on your phone** work best. Warm bulbs change the color of your product: here the white panels came back cream. Warm lightResultNeutral lightResult The flash on your phone points straight at the piece and leaves no shadows: Light from the sideFlash ## Details Shoot a **close up** of any small detail you care about, like stitching or a logo. A flat lay works great for close ups. WaistbandCollar The waistband has three rows of stitching. Without the close up, it came back with two: With the close up, it got all three: ## Putting it together Attach the model image with your product photos and close ups. ModelFrontSideBackWaistbandCollar New conversation > Put this jacket on him. Create three separate images: a front, a side and a back view. FrontSideBack All three views keep the stripes, the colors, and the logo. Even the word on the back of the collar is there. Ask for **three separate images** to get each view as its own picture. ## What to remember - **Views:** the front and the back at least. A side photo is a bonus. - **Fit:** on a person is best, then a mannequin, then a flat lay. - **Light:** neutral light or your phone's flash pointed straight at the piece. - **Details:** close ups of anything small you care about. ## Keep going - Ready to dress a model? Read [Getting Started with Virtual Try-On](/using-fashn/guides/virtual-try-on/getting-started-with-virtual-try-on). - Turn your photos into a whole photo shoot with [Mannequin to Photo Shoot](/using-fashn/guides/agent-workflows/mannequin-to-photo-shoot). - Need a model first? Read [How to Create a Realistic AI Model](/using-fashn/guides/model-creation/how-to-create-a-realistic-ai-model). --- --- title: Product Photos description: Prepare product photos for accurate, detailed results. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Product Photos Prepare product photos for accurate, detailed results. Good product photos give Agent everything it needs to get your product right. This guide shows you how to take them. --- --- title: Prompting in FASHN description: Write clear prompts for consistent results. Start with the basics, then apply tool-specific tips. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Prompting in FASHN Write clear prompts for consistent results. Start with the basics, then apply tool-specific tips. Learn what FASHN handles automatically, where to focus your prompts, and how to prompt each tool effectively. --- --- title: Prompting Fundamentals description: What FASHN handles automatically and where to focus your prompts. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Prompting Fundamentals What FASHN handles automatically and where to focus your prompts. This guide helps you describe the result you want when using FASHN tools that accept text prompts, such as Product to Model, Edit, Packshot, etc. FASHN already keeps images realistic and faithful to your product, so you do not need to say things like “realistic photo” or “copy the product exactly.” Instead, focus your prompt on: ## Focus your prompt on - **Who** is in the shot - **How** they are framed - **Where** the scene takes place Keep it simple - The shorter the better --- ## What FASHN already handles for you You can assume that: - The product stays consistent across images in the same session. - The garment details stay true to the original upload. - The model, garment, and background are rendered in a realistic way. ## You don’t need to include So you do **not** need to repeat generic phrases like: - “realistic”, “hyper-realistic”, “highly detailed” - “copy the product exactly” - “keep the garment the same” ## What you control Use your prompt for the **creative decisions** you control: - Who is wearing the product - How they are posed and framed - Where the photoshoot takes place - Any issues you want to avoid, [phrased as positive instructions](/using-fashn/guides/prompting-in-fashn/writing-effective-prompts#guiding-away-from-unwanted-details) Ready to dive deeper? Continue to [Writing Effective Prompts](/using-fashn/guides/prompting-in-fashn/writing-effective-prompts). --- --- title: Tool-Specific Tips description: Targeted prompting advice for Product to Model, Edit, Packshot, and other Studio tools. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Tool-Specific Tips Targeted prompting advice for Product to Model, Edit, Packshot, and other Studio tools. Apply the best practices from [Writing Effective Prompts](/using-fashn/guides/prompting-in-fashn/writing-effective-prompts) to each tool. Emphasize different parts of the prompt depending on what the tool controls. New to prompting? Start with [Prompting Fundamentals](/using-fashn/guides/prompting-in-fashn/prompting-fundamentals). ### Product to Model Focus your prompt on: - Model description (age range, body type, hair). - Pose and framing (full body vs upper body). - Background, lighting, and overall mood. For your first runs with Product to Model, try generating without any text prompt. The default image can already be close to what you need, or it can give you a clear starting point for deciding what you want to change in the next prompt. You usually **do not need to describe the product** in this tool. The tool reads the product from your upload and keeps it consistent. Mention the product only when: - You want the model to interact with it in a specific way (for example, “jacket worn open over the top”, “scarf wrapped once around the neck”). - You want to fix something the first generation missed or changed (for example, texture, seams, or other small details). **Examples:** - “Full body studio photo of a young adult woman with an average build and shoulder-length wavy hair, standing relaxed with one leg slightly forward, neutral light grey background, soft even studio lighting.” - “Three-quarter view of a young adult man with an athletic build and short curly hair, standing in a modern living room with a light sofa and a large window, soft afternoon daylight.” **If the first result alters the garment**, adjust the prompt by adding **only the missing or incorrect detail**, such as: - “dense, chunky terry fabric with a pronounced, bumpy weave” for a bathrobe that rendered too smooth. - “seamless waistband and smooth side panels” if extra seams or stitching appeared in leggings. **To avoid describing the whole scene from scratch**, you can also: - Use an **Image Prompt** that already has the kind of framing, background, or styling you like. - Then add a short text prompt that focuses only on the model and any key adjustments. ### Model Swap Focus your prompt on detailed facial and skin features. [Model Swap](/using-fashn/studio/model-swap) is all about the person, so you can be specific about: - Hair texture and style. - Eye color and shape. - Skin tone, freckles, beauty spots, wrinkles. - Face shape, jawline, chin, lips, nose. Model Swap can change background and pose while swapping identity, but it works best when you treat it primarily as a **who** change. For more controlled pose or background adjustments, use [Edit](/using-fashn/studio/edit) after you are happy with the new person. Instead of broad labels like “black” or “white”, use more specific national or regional descriptors that tend to produce more recognizable and consistent looks, such as Thai, German, Italian, Indian, Mongolian, Sudanese, or Kenyan. **Examples of identity descriptions:** - “Young adult Thai woman with warm medium brown skin, straight black hair in a low ponytail, soft freckles across her nose, and natural makeup.” - “Middle-aged Italian man with light olive skin, short salt and pepper hair, defined jawline, light stubble, and brown eyes with fine smile lines.” - “Young adult Kenyan man with deep brown skin, buzz cut black hair, high cheekbones, a small gap between his front teeth, and a relaxed friendly expression.” ### Edit [Edit](/using-fashn/studio/edit) is the most versatile tool. It starts from an existing image (usually from Product to Model or Model Swap) and assumes: - You want to keep the same person and product. - You want to make a relatively small, focused change. **You will usually get the best results if you:** - Change one thing at a time, such as pose, arm placement, facial expression, head angle, or a small accessory adjustment. **Examples of single focused edits:** - “Relax the pose so one hand rests lightly in a pocket while the other arm hangs naturally by the side.” - “Tilt the head slightly toward the camera and add a soft, natural smile.” - “Add black sunglasses to the model’s face.” **For larger changes**, it can be more reliable to split the edit into two or more steps. For example: - To go from a direct front view to a full side profile, first ask for a three-quarter view and then, from that result, ask for a side profile. ### Create Model [Create Model](/using-fashn/studio/create-model) builds a new model from scratch. It is often too ambitious to get the exact face, body, outfit, composition, and background all in one prompt, so treat the first generation as a strong starting point. You can always use Edit afterward to adjust expression, accessories, background, or minor pose details. **Use your prompt here to lock in the parts that are harder to change later:** - Body build and proportions. - Overall composition, such as full body vs three-quarter framing and camera distance. - Face shape **Examples:** - “Young adult woman with a curvy figure, medium brown skin with warm undertones, shoulder-length coily hair, relaxed and friendly expression, standing in a full body studio composition on a light grey background.” - “Middle-aged man with an athletic build, fair skin with light freckles, short straight hair with some grey, confident but approachable expression, framed in a three-quarter view from mid-thigh upward.” Create Model also supports an Image Reference, which can ground the composition and pose. **A common pattern is:** - Choose an Image Reference that already has the framing, pose, and camera distance you like. - Then use a short prompt that focuses on body build, age range, and facial features so the model matches your intended look. If you know this model will wear certain items later in Product to Model or Try On, pick a look and composition that will complement those garments, such as a full body standing pose for long dresses or a tighter portrait framing for accessories and beauty shots. ### Try-On [Try-On](/using-fashn/studio/try-on) places a specific garment onto an existing model photo. Unlike Product to Model, the model is fixed and only the clothing changes. **Focus your prompt on:** - How the garment should be worn: "jacket worn open," "shirt tucked in," "sleeves rolled up." - Styling adjustments that help the product look natural on the chosen model. **You usually do not need to describe the model.** Try-On preserves the model's identity, pose, and background automatically. Mention the model only if: - You want the original clothing removed first (for example, "remove the jacket"). - You need to override the default category detection (for example, set Category to "Top" instead of "Auto"). **Examples:** - "Jacket worn open, hands at sides" - "Shirt tucked in with belt visible" - No prompt at all (often produces the most natural result) ### Packshot [Packshot](/using-fashn/studio/packshot) turns a product or on-model photo into a clean catalog packshot. The default is ghost mannequin on a neutral background; the prompt is what you use to pick a different presentation. **Focus your prompt on:** - Presentation style: "flat-lay packshot", "on-hanger", "floating garment". - Surface and lighting: "marble surface", "window-light shadows", "raw concrete floor". - Camera angle when it matters: "top-down view", "three-quarter angle". - Item isolation when the source has multiple pieces: "only the jacket", "just the shoes". **You usually do not need to describe** the product, since it carries over from the source. You also do not need to ask for cleanup, wrinkles, twisted collars, and asymmetric drape are removed automatically. **Examples:** - "Flat-lay packshot on marble with soft window-light shadows" - "On-hanger against a plaster wall" - "Just the shoes, top-down view" - No prompt (defaults to ghost mannequin on a clean background) Easier to show than to describe? Use Image Context to add your own style reference or choose an [Element from Clothing, Accessories, or Shoes](/using-fashn/elements), and keep the prompt short. ### Image to Video [Image to Video](/using-fashn/studio/image-to-video) adds subtle motion to a finished image. Less is more with video prompts. **Tips:** - Running **without a prompt** often produces the most natural, subtle motion. The AI adds gentle camera movement and fabric sway by default. - If you add a prompt, keep it short and specific: "touch fabric," "subtle camera pan," "wind blows hair gently." - Avoid complex action sequences. The tool works best with small, natural movements. **Examples:** - No prompt (recommended for most cases) - "Model turns slightly to the right" - "Gentle breeze moves the fabric" ### Iterate with Agent [Agent](/using-fashn/agent) lets you continue from a result in the same conversation. Describe what you want in natural language, review the result, then give a follow-up instruction for the next version. --- --- title: Writing Effective Prompts description: Core principles and practical tips for writing clear prompts in FASHN. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Writing Effective Prompts Core principles and practical tips for writing clear prompts in FASHN. This guide builds on [Prompting Fundamentals](/using-fashn/guides/prompting-in-fashn/prompting-fundamentals). It focuses on phrasing instructions clearly, iterating effectively, reinforcing key details, and steering away from unwanted results. New to prompting? Start with [Prompting Fundamentals](/using-fashn/guides/prompting-in-fashn/prompting-fundamentals) to learn what FASHN handles automatically and where to focus your prompts. ## Core principles for good prompts These principles help your prompts stay clear, consistent, and easy to refine. ### Use natural, plain language Write prompts as you would brief a photographer: - Use full sentences instead of short comma-separated tags: “full body studio photo of a young adult woman” instead of “studio, woman, full body”. - You do not need special syntax, weights, or flags. ### Put the most important information first Models tend to pay more attention to the start of the prompt. Lead with: 1. The type of shot and subject. 2. The model’s key traits. 3. Background and mood. Example: > Full body studio photo of a young adult woman wearing the outfit, average build, neutral light grey background, soft even lighting. ### Use concrete, visual language Replace adjectives like “beautiful” or “cool” with concrete visual details: **Instead of vague phrases, try:** - “beautiful background” → “simple light grey studio background” or “modern living room with a light sofa and a large window” - “cool vibe” → “relaxed pose, slight smile, soft natural lighting” ### Use enough detail, but keep prompts focused Longer prompts can give more control, but very long prompts can cause the model to ignore the end. Aim for a **small number of clear sentences** (often 1–3), not a paragraph full of loosely related ideas. --- ## A simple prompt structure for FASHN You can think of your prompt as five parts. You do not need to mention every part every time, but this structure is a reliable starting point. **Shot summary.** One sentence that describes the overall image type and subject. Examples: - Full body studio photo of a young adult woman wearing the selected outfit. - Head-and-shoulders portrait of a young adult man wearing the selected outfit. **Model details.** A neutral description of the model’s age range, body type, hair, and overall vibe. Examples: - Average build, shoulder-length curly hair, soft natural makeup, relaxed and confident expression. - Plus-size model, short coily hair, warm friendly expression. - Athletic build, short straight hair, focused expression. **Pose and framing.** What the model is doing and how the camera sees them. Helpful terms: - Framing: full body, three-quarter view, head-and-shoulders portrait, knee-high shot. - Angle: eye-level, high angle, low angle, three-quarter view, side profile. Examples: - Full body front view, standing straight with one leg slightly forward, arms relaxed by the sides. - Three-quarter view, one hand in a pocket, gentle smile, eye-level camera. **Background and setting.** Where the photoshoot is happening. Examples: - Clean studio background in light grey with a soft floor shadow under the feet. - Minimal bedroom with white bedding and a large window. - City sidewalk with soft focus buildings in the background. **Lighting and mood.** How the shot feels. Examples: - Soft even studio lighting, calm and polished mood. - Warm afternoon sunlight from the side, relaxed lifestyle feel. - Bright, neutral lighting, e-commerce catalog style. You can often omit one or two parts and still get good results, but keeping this order makes prompts easier to write and refine. --- ## Guiding away from unwanted details Text-to-image models often struggle with words like “no”, “not”, or “without”. If you write: > a woman without jewelry the model may still focus on the word “jewelry”. The same goes for phrases like “no pockets”, "without stitches”, or "belt not visible". The most reliable workaround is to describe the **positive visual opposite** of what you want to exclude. ### Rewrite negatives as positives If something keeps appearing that you do not want, try: - Removing negative phrases like “no” or “without”. - Adding a short positive phrase that describes what is visible instead. **Instead of:** “no people in the background” → “an empty studio background” or “a plain light wall behind the model” “no jewelry” → “bare neck and bare hands, clear skin” “no pockets in the leggings” → “leggings with clean and simple side panels” “jeans with no belt” → “jeans with a metal tack button visible on the front” ### Reinforce key details Rephrasing key details in natural variations helps the model stay focused: - “full body photo of the model” and later: "visible from head to toe, with white sneakers and shoulder-length wavy hair fully in frame". - “waistband of the pants clearly visible” and later: “the top is neatly tucked into the pants so the entire waistband and front belt loops are visible in the shot”. Use repetition sparingly to reinforce the core of the shot, and only when a single instruction doesn’t guide the model reliably. --- ## Describing models clearly You can guide the model’s appearance without specifying exact ages or measurements. Use clear, consistent language that describes what should be visible in the photo. These patterns help you describe models clearly and consistently. ### Age and life stage Use life-stage terms rather than specific numbers, unless you genuinely need a number: - Young child, older child, teenager. - Young adult, adult, middle-aged adult. - Older adult, senior. You can add gentle modifiers for tone: - “young adult woman, fresh-faced, minimal makeup”. - “middle-aged man with a composed expression”. - “older adult woman with silver hair and a warm smile”. ### Body type Use descriptive, visually focused phrasing: - Slim build, average build, athletic build. - Curvy figure, plus-size model. - Stocky build, tall and lean. You can add simple visual cues: - “broad shoulders, long legs, athletic build”. - “soft curves, plus-size model, confident posture”. ### Skin tone and color Use clear, consistent terms: - Light skin, medium brown skin, deep brown skin. - Warm olive tone, soft caramel tone, rich espresso tone. If you mention ethnicity, pair it with a simple skin tone description: - “South Asian woman with warm brown skin”. - “Northern European man with fair skin and freckles”. ### Angle of view and perspective Terms that help control how the model is seen: - Full body shot, waist-up portrait, knee-high shot. - Front view, three-quarter view, side profile, back view. - Eye-level, low angle, high angle, over-the-shoulder. Example: > Full body front view of a young adult woman with an average build and medium brown skin, eye-level camera, relaxed posture, soft natural expression. --- ## Iterating and refining your prompt It is normal if the first result is not exactly what you imagined. Iteration is how you get from “close enough” to “just right”. ### Change one thing at a time When you refine a prompt, pick **one** aspect to adjust: - Model description (age range, body type, hairstyle). - Outfit (fit, accessories, details). - Pose or framing (full body, upper body, side profile, back view). - Background (studio, bedroom, outdoor). - Lighting (soft, natural light, bright). This makes it easier to see what worked: - Original: “full body studio photo, neutral light grey background”. - Iteration: “full body studio photo, pure white background”. ### Start simple, then add detail You can build a prompt in stages: 1. “Full body photo of a young adult woman.” 2. “Full body photo of a young adult woman with an average build and shoulder-length wavy hair.” 3. “Full body photo of a young adult woman with an average build and shoulder-length wavy hair, standing in front of a simple light grey backdrop with soft, even lighting.” 4. “Full body photo of a young adult woman with an average build and shoulder-length wavy hair, standing in front of a simple light grey backdrop with soft, even lighting, relaxed confident expression and one leg slightly forward.” Small, controlled changes make it easier to steer the results. ### Try alternate wording If a phrase is not giving you what you expect, try a close synonym that is more visual: **Examples:** - “street style fashion photo” → “outdoor fashion photo on a city sidewalk”. - “cozy room” → “small bedroom with soft blankets and warm lamp light”. - “modern background” → “minimal living room with a light sofa and a large window”. Next up: apply these patterns in specific tools. See [Tool-Specific Tips](/using-fashn/guides/prompting-in-fashn/tool-specific-tips). --- --- title: Getting Started with Virtual Try-On description: Put any product on any model in Agent, from product photos and clothing types to accessories, poses, and angles, with the exact prompts from the video. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Getting Started with Virtual Try-On Put any product on any model in Agent, from product photos and clothing types to accessories, poses, and angles, with the exact prompts from the video. Virtual try-on puts your product on a model. You start with a **model image** and a **product image**, and the images do the work instead of a long prompt. The model stays exactly the same and the garment comes out just like your product photo. This guide follows the video step by step in [Agent](/using-fashn/agent), from a single dress to a full outfit with accessories, a new pose, and a back view. Every example below was made in Agent at **2K**, **Balanced** (set in [Generation settings](/using-fashn/agent/generation-settings)). ## What you need ### Model image The person who will wear your product. Use a model you [created in Agent](/using-fashn/guides/model-creation/how-to-create-a-realistic-ai-model), a model from [Elements](/using-fashn/elements), or any photo of a person. ### Product image Product photos come in three types: on-person, flat-lay, and mannequin. All three work, but on-person photos are the best because they show exactly how the product should fit. On-personFlat-layMannequin ## Product photo ### On-person Attach the model image and the product image, then send the message. New conversation > Put the product on the model ModelProductResult The result looks like the model image except for the dress, which is the only thing taken from the product photo. ### Flat-lay A flat-lay on the floor works just as well, and it is the easiest product photo to take. This time, try it without a text prompt. New conversation No text prompt ModelProduct Agent asks what you want to do. Continue the conversation > Virtual try-on ModelProductResult ### Mannequin New conversation > Put this product on this model ModelProductResult All three product photos work great. Whichever you use, make sure the details you care about are visible in the product image. ## Clothing types ### Top New conversation > Put the product on the model ModelProductResult Only the top changed. The jeans, shoes, pose, background, and model stayed the same. ### Bottom New conversation > Put the skirt on her ModelProductResult ### Fix with text This skirt is meant to be low-waisted, but it came back high-waisted. Use text to tell Agent the change you want to make. Continue the conversation > The skirt should be low-waisted BeforeAfter ### Top and bottom You can send more than one garment at a time, with up to [four images](/using-fashn/agent/create-and-edit-images#work-with-multiple-images) in one message. New conversation > Put these products on the model ModelTopBottomResult ### Jacket New conversation > Put the jacket on her ModelProductResult Jackets and outerwear go over what the model is already wearing instead of replacing it. ## Accessories Accessories work the same way. Upload your own accessories or choose from [Elements](/using-fashn/elements). New conversation > Put these on her ModelBootsBagResult The boots and bag were added to the look and everything else stayed the same. ## Pose and angle You can shoot the finished look in different poses and from different angles. ### Change the pose New conversation > Change her pose BeforeNew pose Only the pose changed and everything else stayed the same. You can also be specific, for example "she is walking towards the camera" or "her pose is relaxed." ### Show the back Reference the original image by selecting its [**@** button](/using-fashn/agent/reference-images) in the media panel and describe the change. Continue the conversation > @input1 show me the back FrontBack If the back does not look right, add a photo of the back or fix it with text. Side views and close ups work the same way. ## What to remember - **What you need:** a model image and a product image. - **Best product photo:** on-person. Flat-lay and mannequin photos work too. - **Fit looks off:** describe the fix in the same conversation. - **Several pieces:** send them together in one message. - **More angles:** reference the original image, not the last result. ## Keep going - Need a model first? Read [How to Create a Realistic AI Model](/using-fashn/guides/model-creation/how-to-create-a-realistic-ai-model). --- --- title: Virtual Try-On description: Put your products on a model image you already have, and change only the garment. --- > Documentation Index > Fetch the complete documentation index at: https://help.fashn.ai/llms.txt > Use this file to discover all available pages before exploring further. # Virtual Try-On Put your products on a model image you already have, and change only the garment. Virtual try-on keeps your model exactly as they are and changes only what they wear. All you need is a model image and a product image.