开发者文档

neo-fashion.ai API

Generate product photography, video, and campaign imagery programmatically. The same API the platform uses — available to your integration from day one.

Base URL

All API requests use the following base URL. There is no versioning beyond v1 — all current endpoints are stable. Breaking changes will be released under v2 with a minimum 6-month deprecation window.

Base URL

https://app.neofashion.ai/api/v1

Authentication

The API accepts two authentication strategies on every endpoint. Both resolve to a workspace and user context server-side.

API Key

For server-to-server integrations. Generate API keys in your workspace settings. Keys are prefixed ne_.

Authorization: Bearer ne_live_xxxxxxxxxxxx

JWT (Session)

For web UI sessions. Supabase Auth JWT is accepted on every endpoint — the same path the platform frontend uses.

Authorization: Bearer eyJhbGci...

All requests include a source field (ui or api) and optional api_key_id for audit and billing purposes.

Core Endpoints

Generation endpoints consume credits from your workspace balance before calling the AI provider. Credits are refunded automatically if the provider returns an error.

Method	Endpoint	Description
POST	`/api/v1/generate/image`	Generate a single product photo
POST	`/api/v1/generate/video`	Generate a short fashion video
POST	`/api/v1/generate/bulk`	Async batch generation job
POST	`/api/v1/sketch-to-photo`	Sketch → campaign-quality photo
GET	`/api/v1/generations`	List generations (paginated)
GET	`/api/v1/generations/:id`	Generation detail + status
GET	`/api/v1/credits/balance`	Current credit balance
GET	`/api/v1/personas`	Workspace personas
GET	`/api/v1/models`	Custom trained models

Response Shape

Every endpoint returns the same envelope. This makes error handling consistent across all operations regardless of the generation type.

Success (200)

{
  "data": {
    "id": "gen_01hwz...",
    "status": "completed",
    "output_url": "https://...",
    "credits_used": 50
  },
  "meta": {
    "credits_used": 50,
    "credits_remaining": 7950,
    "request_id": "req_01hwz..."
  },
  "error": null
}

Error (4xx / 5xx)

{
  "data": null,
  "meta": { "request_id": "req_01hwz..." },
  "error": {
    "code": "INSUFFICIENT_CREDITS",
    "message": "Credit balance too low for this operation.",
    "details": { "required": 50, "available": 20 }
  }
}

Rate Limits & Plans

Rate limits apply per workspace. Limits reset on a rolling 60-second window. The current limit is returned in every response via X-RateLimit-Remaining and X-RateLimit-Reset headers.

Plan	Rate limit
Starter	`60 req / min`
Pro	`300 req / min`
Enterprise	`Custom SLA`

Error Codes

Structured error codes are always present in the error.code field. HTTP status codes follow standard REST conventions.

Code	HTTP	Meaning
`INSUFFICIENT_CREDITS`	402	Workspace credit balance too low for this operation
`UNAUTHORIZED`	401	Missing or invalid API key / JWT
`FORBIDDEN`	403	Feature not available on current plan
`NOT_FOUND`	404	Generation or resource does not exist in this workspace
`RATE_LIMITED`	429	Request rate exceeds plan limit
`PROVIDER_ERROR`	502	AI provider returned an error; credits refunded automatically

Get started

Ready to integrate?

API access is available on Pro and Enterprise plans. Create a workspace and generate your first API key in under two minutes.

Book a demo View pricing