Product · API

Fashion image generation API for enterprise production.

Use Uwear as the developer access layer for structured product assets, async generation jobs, QA, and delivery into your DAM, PIM, storefront, agent workflow, or internal tools.

Book a demo Read API docs

create-generation.http

POST

POST /generation
Authorization: Bearer YOUR_API_KEY

{
  "clothing_item_ids": [123, 124],
  "model_slug": "nano-banana-2",
  "use_case": "generate",
  "prompt": "full body studio ecommerce image",
  "camera": "full_body_front",
  "aspect_ratio": "9:16",
  "resolution": "2K",
  "num_images": 2,
  "art_direction_id": 501,
  "do_qa": true,
  "max_qa_retries": 1,
  "webhook_events": [
    "generation.completed",
    "generation.failed"
  ]
}

Flat clothing item sent to the Uwear API

Second flat clothing item sent to the Uwear API

Generated on-model result returned by the Uwear API for the two clothing items

Base URL: https://api.uwear.ai
Auth: Bearer API keys
Format: snake_case JSON
Completion: Polling or webhooks

One platform, four ways in

The same production infrastructure, reached your way.

Studio, API, Agent Mode, and the MCP share one model library, one clothing workspace, and one generation history. Pick the access path that fits the work, and move between them as the job changes.

Programmatic access
API
A REST API for async generation, edits, video, and identity. Wire production workflows into your catalog, DAM, and storefront.
You are here
Operator interface
Studio
The visual workspace for creative production. Batch catalogs, consistent models, edits, backdrops, video, and upscaling in one surface.
Explore
Conversational orchestration
Agent Mode
Describe a shoot in plain language and Mouza builds the brief, assigns models, styles outfits, and runs the batch end to end.
Explore
In your own tools
MCP
Connect Uwear to Claude, ChatGPT, or any MCP client and run the same production from the tools your team already uses.
Explore

System context

The API keeps production structure attached to every image.

A technical integration should not collapse fashion production into a prompt string. Uwear exposes the objects your systems need to track: product inputs, reusable creative direction, generation jobs, QA state, and delivery metadata.

Product inputs

Create clothing items, outfits, avatars, locations, uploaded files, and tags before the generation run. Product records can carry source metadata for catalog and DAM traceability.

Explore asset library

Creative controls

Apply art_direction_id, location_id, reference attachments, camera, aspect ratio, resolution, model_slug, seed, and advanced model parameters when a workflow needs them.

Explore art direction

Async generation jobs

Create images, edits, upscales, and videos through POST /generation. Use children to chain follow-up steps after parent images complete.

Explore batch workflows

Result records

Retrieve generation_results with output URLs, thumbnails, result kind, origin, source IDs, QA status, tags, semantic search filters, and download paths.

Explore QA and retries

Async contract

Create a job, then let your system wait for completion.

The API uses one mental model across on-model images, edits, upscales, and videos: create the job, monitor it, then deliver the generated result into the system that owns the customer experience.

Read API docs Download UWEAR.md

generation-status.json

POLL

GET /generation/789

{
  "generation_id": 789,
  "status": "Done",
  "generation_results": [
    {
      "generation_result_id": 1234,
      "url": "https://...",
      "thumbnail_url": "https://...",
      "kind": "Image",
      "qa_status": "completed",
      "qa_decision": true,
      "clothing_item_id": 123,
      "art_direction_id": 501,
      "origin": "api"
    }
  ]
}

request-webhooks.json

WEBHOOK

POST /generation  // excerpt

{
  "webhook_url": "https://brand.com/uwear/events",
  "webhook_secret": "whsec_...",
  "webhook_events": [
    "generation.completed",
    "generation.failed"
  ]
}

// Or save an endpoint once on the Webhooks page.
// Terminal payloads put assets in data.results.

How it works

From product manifest to approved asset record.

Create or map source assets

Store the Uwear clothing_item_id, outfit_id, avatar_id, location_id, or art_direction_id beside the product, variant, model, campaign, or workflow record in your own system.

Submit one structured job per output need

Send a generation request with explicit model, use case, camera, resolution, aspect ratio, references, QA settings, and optional webhook configuration.

Wait without blocking the user

Poll GET /generation/{generation_id} for scripts and internal tools, or configure webhooks for production pipelines that need terminal callbacks.

Persist the approved result

Read generation_results, inspect QA state, download assets while URLs are available, then store final images, videos, IDs, and metadata in your DAM, PIM, CMS, app, or queue.

Uwear batch builder with selected inputs, generation settings, CSV batch upload, and completed runs

Batch and integration

Run catalog work as manifests, batches, or agent jobs.

Enterprise API work usually starts with a product source of truth. Uwear fits that shape: each row can map to a clothing item, outfit, model, art direction, location, camera, output format, QA rule, and delivery destination.

catalog-row.json

MANIFEST

{
  "sku": "BRN-STRIPE-012",
  "clothing_item_id": 123,
  "avatar_id": 101,
  "art_direction_id": 501,
  "location_id": 77,
  "model_slug": "nano-banana-2",
  "camera": "full_body_front",
  "destination": {
    "dam_folder": "spring-catalog",
    "pim_field": "on_model_image"
  }
}

Operating modes

The same API supports several enterprise integration paths.

API access sits behind the workflow your team already owns, whichever shape that takes.

Catalog pipeline

Turn a product manifest into controlled generation jobs, track each generation_id, and only rerun failed or rejected rows.

Storefront or app backend

Keep API keys on your server, accept shopper or product context from the frontend, then deliver the completed result back to the experience.

Creative production tooling

Embed Uwear in internal tools where operators choose models, art direction, locations, references, QA rules, and final delivery destinations.

Agent workflows

Give your coding agent UWEAR.md so it can plan jobs, poll statuses, handle webhooks, preserve result metadata, and avoid unsupported routes.

Result delivery

Generated assets come back with the context operations need.

The handoff is not just an image URL. Result records carry enough context to filter, QA, retry, download, and map final assets back to the product or campaign that requested them.

generation_result_id, generation_id, kind, thumbnail_url, and output url
source IDs for clothing items, outfits, avatars, locations, art directions, and models
qa_status and qa_decision for review, filtering, retries, and delivery gates
tag filters, semantic search, date filters, origin filters, and batch filters
download ZIP routes for selected approved outputs

Generated asset URLs are temporary delivery URLs. Production integrations should copy approved files into their own storage before they expire.

Flat product input used by the Uwear API before generating an on-model visual

Generated on-model fashion visual returned as a Uwear generation result

Metadata traceability

Metadata in. Metadata out.

Attach your own metadata to clothing items and models: SKUs, style numbers, campaigns, markets, usage rights, DAM or PIM identifiers. It travels through generation untouched, so an exported image can always answer where it came from and what it may be used for.

Set it once on the clothing item or model: up to 50 fields of your own keys.
Every future generation inherits the source metadata automatically.
Every result hands it back, embedded in the record your systems read.

clothing-item.json

POST /clothing-item  // excerpt

{
  "name": "Vintage denim jacket",
  "metadata": {
    "sku": "DJ-2026-BLU",
    "style_number": "ST-1042",
    "season": "Spring 2026",
    "campaign": "Spring ecommerce refresh",
    "market_region": "US",
    "usage_rights": "Ecommerce and paid social"
  }
}

generation-result.json

OUT

GET /generation-result/1234  // excerpt

{
  "generation_result_id": 1234,
  "clothing_item_id": 123,
  "embedded_metadata": {
    "uwear-generation": {
      "generation_id": 789,
      "clothing": [{
        "id": 123,
        "name": "Vintage denim jacket",
        "metadata": {
          "sku": "DJ-2026-BLU",
          "campaign": "Spring ecommerce refresh",
          ...
        }
      }]
    }
  }
}

Related workflows

Connect API access to the rest of the production system.

See IT evaluation path

Batch generationRun catalog jobs from structured production inputs.Learn more Asset libraryManage clothing, outfits, avatars, locations, files, and tags.Learn more QA and retriesReview generated outputs and retry only the exceptions.Learn more Agent ModeUse an agent to assemble the production brief and run.Learn more IT and API teamsEvaluate security, throughput, integration, and operations.Learn more AI fashion API guideGive a coding agent the API workflow and UWEAR.md reference.Learn more

Map Uwear into your production stack.

Bring your catalog source, integration pattern, QA requirements, and delivery target. We will help define the API path that fits.

Book a demo Open Uwear

Existing customer? Log in to Studio