About LLMCompat

LLMCompat helps developers locate failures in OpenAI-compatible APIs. It tests the endpoint and records evidence for each conclusion.

Focus
API compatibility diagnosis
Method
Requests, assertions, evidence

Why LLMCompat Exists

A successful chat response does not prove API compatibility. LLMCompat tests separate contracts so developers can identify the failed layer.

How We Create Guides

Each guide starts with a specific integration problem. We compare the expected protocol with observable endpoint behavior and provide a verification step.

What Counts as Evidence

  • Raw request and response shapes from the current probe suite.
  • Assertions that state the expected and actual values.
  • Reproduction commands that a developer can run independently.
  • Primary technical sources for external protocol requirements.

Credential and Report Privacy

LLMCompat does not store API keys or reports. Read the full credential policy and test methodology.