Ada AI Documentation Center
Integrate Ada AI with confidence.
Explore product guides, API references, and integration examples for the Ada AI model gateway. One endpoint, every provider, full control.
Browse by topic
Documentation
Quickstart
Send your first request in five minutes: sign in, add an upstream, mint a key, and route traffic.
Chat completions
The OpenAI-compatible endpoint — streaming, tool calls, and SDK examples in Python, JavaScript, and the Vercel AI SDK.
Anthropic messages
Use /v1/messageswith the Anthropic SDK to talk to any provider's model through one shape.
Models & routing
How model discovery, priority ordering, and automatic 5xx failover work across your upstreams.
API keys
Mint proxy keys, scope them to yourself or an organization, and rotate or revoke them at any time.
Organizations
Share a pool of upstreams with your team under one account, with roles and invitations.
Coding agents
Point Pi, OpenCode, and other agents at Ada AI so every request flows through your proxy key and routing rules.
Usage & billing
Track tokens, latency, time-to-first-token, provider, and status via the dashboard and API.
From the blog
Latest updates
How automatic failover works
A look at how Ada AI retries upstreams when a provider returns a 5xx, without changing your client code.
Announcing Ada AI
One OpenAI-compatible endpoint across every provider you already use. Bring your keys, keep control.
View all postsRead the latest updates from the team.Go to blog