Skip to main content

​
Getting started

How do I send my first request? Follow the Quickstart — it takes you from sign-up to a traced LLM call. What’s the difference between an organization, a workspace, and a project? Everything nests into Organization → Workspace → Project. See Core Concepts. Do I need to change my application code? No. Point your SDK’s base URL at the gateway and authenticate to to11 with the x-to11-authorization header — your application logic stays the same. See Integrations.

​
Providers & integrations

Which SDKs are supported? The OpenAI, Anthropic, and Vercel AI SDKs. See Integrations. Which model providers can I use? OpenAI, Anthropic, and Cohere have native adapters. Many more work through the gateway as OpenAI-compatible providers (such as xAI, Mistral, and Groq) or as cloud platforms with their own auth (Azure OpenAI, Google Gemini, and AWS Bedrock). See Supported providers for the full list. Can I send a request in one format and route it to a different provider? Yes. The gateway translates between wire formats, so you can send an OpenAI-format request and route it to Anthropic (and back). See Deploy.

​
Observability & data

Where do I see my traces? Open your project and go to Traces. Every call through the gateway becomes an OpenTelemetry-native trace. See Observe. Does to11 record my prompts and completions? Yes. Content capture is on by default, and prompt and completion text is recorded verbatim on your spans. It’s a gateway-level setting managed by to11 — there’s no per-request header or dashboard toggle to turn it off. If your data-handling requirements mean this content must not be stored, contact to11. See Content capture. Is evaluation available? Not yet — it’s coming soon. See Evaluate.

​
Access & security

How do I invite a teammate? See Team. How do I control who can do what? See Roles & permissions. How do I create or rotate an API key? See API keys.

​
Still stuck?

Where do I report a bug or ask a question? The to11 community on GitHub is the place to ask questions, report bugs, and request features. See Getting help.