Skip to content
Last updated

The Image Manipulation API provides AI-powered image processing operations using multiple providers (Clipdrop and OpenAI).

Base URL: /api/v1/images

Supported Operations

  • Remove Background - Extract subjects with transparent backgrounds
  • Remove Text - Remove text overlays from images
  • Cleanup - Remove or replace objects using mask-based editing
  • Replace Background - Replace backgrounds with AI-generated scenes
  • Generate - Create images from text prompts
  • Uncrop/Outpaint - Extend images beyond original boundaries
  • Upscale - Enhance resolution while maintaining quality

Providers

The API supports two image processing providers:

  • Clipdrop - Specialized in image manipulation operations
  • OpenAI - Supports generation and mask-based editing

Each operation may support one or both providers. Provider-specific requirements are documented per endpoint.

Authentication

All endpoints require authentication via Bearer token or API key.

Common Parameters

ParameterTypeRequiredDescription
providerstringYesProvider name: "clipdrop" or "openai"
organization_idstringYesOrganization ID (org_ prefixed string)

Image Formats

Supported Input: PNG, JPEG, WebP

Output Formats: PNG (most operations), JPEG (uncrop, upscale), WebP (replace-background, upscale)

Rate Limiting

  • Clipdrop: 60 requests per minute per organization
  • Response: 429 status code when limit exceeded

Usage Tracking

All successful operations are tracked for billing using a synapse-based system with operation-specific weights.