API documentation

Eclogue exposes one unified API: reach every major AI model through a single endpoint, with key management, usage tracking and rate limiting built in.

Base URL

Every API is served from a single domain:

https://api.eclogue.ai

Main endpoints

Category	Endpoint	Description
Inference	`/v1/chat/completions`	Chat completions, OpenAI format
Inference	`/v1/messages`	Messages, Anthropic format
Inference	`/v1/models/{model}:generateContent`	Vertex / Gemini native protocol
Inference	`/v1/responses`	OpenAI Responses (the newer agent protocol, e.g. GPT-5.1)
Inference	`/v1/videos/generations`	Video generation
Query	`/v1/models`	List available models
Query	`/v1/auth/key`	Check a key's balance and usage
Query	`/v1/generation`	Look up a single request

Key concepts

Several API protocols, one gateway:
- /v1/chat/completions — OpenAI Chat Completions (works with the OpenAI SDK)
- /v1/messages — Anthropic Messages (works with Claude Code and the Anthropic SDK)
- /v1/models/{model}:generateContent — Vertex / Gemini native protocol (works with the @google/genai SDK)
- /v1/responses — OpenAI Responses (the newer agent protocol; works with responses.create())
- /v1/videos/generations — video generation (asynchronous jobs)
Multi-provider routing — automatic load balancing and failover
One API key, balance and billing system across every protocol
Real-time usage tracking and balance management