PicRenew Developer API
The PicRenew API lets you restore and enhance photos programmatically. You can submit restoration jobs, poll their status, and receive results via webhook — all from your own application.
Base URL
Quick start
- Create an API key from your account page (under "Developer API").
- Include the key as a Bearer token in every request.
- Submit a restoration with
POST /v1/restorations. - Poll
GET /v1/restorations/:iduntilstatusis complete. - Download the output from
output_url.
Authentication
All requests require an Authorization: Bearer <key> header. API keys are prefixed with pk_live_. See the Authentication page for full details.
Rate limits
| Scope | Limit | Window |
|---|---|---|
| Per API key | 60 requests | 1 minute (sliding window) |
When the limit is exceeded the API returns 429 Too Many Requests with a Retry-After header (seconds until the window resets).
Error responses
All errors return JSON with an error field:
| Status | Meaning |
|---|---|
400 | Bad request — invalid parameters |
401 | Missing or invalid API key |
402 | Insufficient credits |
404 | Resource not found |
429 | Rate limit exceeded |
500 | Internal server error |
MCP integration
PicRenew also ships an MCP (Model Context Protocol) server at https://picrenew.com/api/mcp — the same Bearer API key authenticates it. See the MCP endpoint page.
Credits
Credit cost is based on output resolution only. All restoration presets (standard, enhance, colorize, portrait) cost the same — there is no surcharge for options.
| Operation | Cost |
|---|---|
| Restore (any preset, any options) | 1 credit (1k) · 2 credits (2k) · 3 credits (4k) |
| Superscale | 1 credit · free for subscribers |
Options like colorize and remove_frame do not add to the credit cost — they change the prompt sent to the AI, not the price.
Credits are debited atomically when the restoration is created. If the job fails to enqueue, credits are automatically refunded.
Next steps
- Authentication — key format, security
- Restorations — full endpoint reference
- MCP endpoint — use PicRenew inside AI clients