Platform · OpenAI

Observe AI agents that call OpenAI.

Your agent calls OpenAI for chat, embeddings, assistants, or the Responses API. Send its OpenTelemetry traces to Trefur and each invocation becomes a span on the agent trace — with the prompt the model saw, the completion it returned, and every tool call in between.

What you see.

Chat Completions — model, prompt, completion, tool calls, stream mode, finish reason.

Responses API — single span per response, tool steps as children.

Embeddings — model, input length, dimensions, tokens, USD attributed.

Assistants — thread ID, run ID, every step on the run, tool outputs submitted.

Streaming completions captured as one span with the full text on close.

Rate limits (429), context-length errors, model deprecation warnings.

Per-org and per-project tagging if you set the OpenAI-Organization header.

Why it matters.

For engineers

One line at startup auto-instruments the OpenAI SDK. Existing code keeps working; the trace shows every call with the prompt the model saw and the completion it returned.

For platform teams

Roll up OpenAI usage per agent, per environment, per project. Give finance a clean breakdown. One trace shape so swapping to Azure or Anthropic later does not break dashboards.

For business operators

See which workflows depend on OpenAI and what they cost. Catch silent model changes (gpt-4o → 4o-mini fallbacks) that drift cost-per-run.

For SRE

OpenAI throttling and context-length errors land on the span. Envelope alerts catch retry-rate spikes and route to oncall with the trace already attached.

How it works.

The OpenAI Admin API connector pulls every project, member, API key, daily token and cost roll-up, and audit event every 15 minutes, with no code change. For prompt and completion content, tokens and tool calls on each call, send your agents' OpenTelemetry traces to Trefur over OTLP/HTTP.

See what your OpenAI agents are actually doing.

Start free. First trace in under five minutes.

Start free See all platforms