Skip to main content
to11 puts an LLM gateway, inline security, and OpenTelemetry-native observability behind one hosted control plane. Route your model traffic through it and to11 traces, secures, and measures every call — without you wiring up a stack of separate tools. It takes your app to eleven.

​
Who it’s for

AI and platform engineers who:
  • Ship LLM-powered applications or agents to production
  • Need observability over non-deterministic model behavior
  • Want inline security — PII detection and blocklist matching applied in the request path, before anything reaches the provider
  • Run more than one provider and want a single interface in front of them

​
What you get

​
Gateway

An LLM proxy between your application and your providers, adding under a millisecond of overhead. Cross-format routing — send OpenAI format, route to Anthropic, get OpenAI back — plus inline security and streaming. See Deploy.

​
Observability

Point your SDK at the gateway and every call becomes an OpenTelemetry-native trace: model, tokens, latency, and optionally the prompt and completion. See Instrument and Observe.

​
Evaluation

Coming soon — score and compare your application against recorded traces. See Evaluate.

​
How it’s organized

Everything nests into Organization → Workspace → Project. See Core Concepts.

​
Next steps

Quickstart

Send your first traced request.

Core concepts

How to11 is organized.