Requesty
requesty.ai
AI gateway and LLM router for 600+ models. Intelligent routing, caching, failover, observability, and enterprise governance through a single OpenAI-compatible API.
Key facts
- Card
- https://www.requesty.ai/.well-known/agent-card.json
- Card version
- A2A v1.0 shape
- Interfaces
- HTTP/REST, protocol version 1.0, at
router.requesty.ai
- HTTP/REST, protocol version 1.0, at
- Streaming
- Yes
- Push notifications
- No
- Extended card
- No
- Authentication
- HTTP Bearer
- Signed
- No
- Provider
- Requesty
- Documentation
- https://docs.requesty.ai/
- Tranco rank
- 152,114
- Last seen
Skills
Chat Completions
POST /v1/chat/completions - Send chat completion requests to 600+ AI models. Supports streaming, function calling, structured outputs, vision, and extended thinking. Docs…
Responses API
POST /v1/responses - OpenAI Responses API for multi-turn conversations with built-in tool use, web search, file search, and computer use. Docs…
Messages API
POST /v1/messages - Anthropic Messages API for Claude models with native support for extended thinking, tool use, and vision. Docs: https://docs.requesty.ai/api-reference/endpoint/messages-create
Intelligent Model Routing
Route requests across 30+ providers by cost, latency, or availability. Weighted load balancing, automatic failover chains, and latency-based routing. Docs…
Embeddings
POST /v1/embeddings - Generate text embeddings using models from OpenAI, Cohere, Voyage, and more. Docs: https://docs.requesty.ai/api-reference/endpoint/embeddings-create
MCP Gateway
Unified gateway connecting AI coding tools (Cursor, Claude Code, VS Code) to any MCP server with auth, analytics, and management. Docs: https://docs.requesty.ai/features/mcp-gateway
Image Generation and Editing
POST /v1/images/generations and /v1/images/edits - Generate and edit images using DALL-E, Stable Diffusion, and other models. Docs…
Audio (TTS and Transcription)
POST /v1/audio/speech for text-to-speech, POST /v1/audio/transcriptions for speech-to-text. Docs: https://docs.requesty.ai/api-reference/endpoint/audio-speech-create
Check this agent
Run the card through the validator, or send the agent a few harmless test calls. Each tool opens with requesty.ai filled in.