everyapi
Integration · LLM ObservabilityOpenAI-compatible · one key · live model catalog
Langfuse

Run the EveryAPI gateway inside Langfuse

Langfuse is an open-source observability platform that can trace application calls made with supported instrumented clients. EveryAPI remains the inference and billing endpoint; Langfuse records the telemetry your application sends to it.

01Setup code

One base URL line points Langfuse at EveryAPI

Default global Base URL / China-first Base URL: https://api.everyapi.ai/v1 / https://api-cn.everyapi.ai/v1

02Steps

From zero to your first call

  1. Sign up for Langfuse

    Sign up at cloud.langfuse.com (SaaS) or self-host via docker compose. Get your public key + secret key.

    cloud.langfuse.com

  2. Install the Langfuse SDK

    pip install langfuse openai or npm i langfuse openai. Langfuse is a wrapper, not a replacement.

  3. Point the OpenAI client at EveryAPI

    Set base_url to https://api.everyapi.ai/v1 and api_key to sk-everyapi-... — same as a normal EveryAPI setup.

  4. Let Langfuse wrap the client

    Use the integration documented for your installed Langfuse SDK version to instrument the OpenAI client that points at EveryAPI. Langfuse SDK APIs change over time, so follow its current official example.

  5. Inspect traces

    Inspect the telemetry captured by your Langfuse instrumentation. Treat the EveryAPI dashboard as authoritative for EveryAPI billing; observability tools may calculate cost using different price metadata.