Helicone is an open-source AI gateway and observability platform: you route model requests through it, and it logs them while giving access to many providers through one API.
In plain words
Instead of adding tracing code everywhere, you change the base URL your OpenAI client points to so requests pass through Helicone. It forwards them to the chosen model, from over 100 across providers including OpenAI, Anthropic and Google, and logs each request automatically. It can fall back to another model when one fails. You can use the hosted service or run it yourself.
A quick example
Imagine a small app already using the OpenAI SDK. The developer changes the base URL to Helicone's gateway and swaps in a Helicone key. From then on every request shows up in a dashboard, and the app can call a Claude or Gemini model by changing the model name.
Where you'll meet it
- Logging and monitoring LLM requests.
- One API for many model providers.
Related
Sources
- DocsHelicone/helicone, Helicone (GitHub) · read 29 Sept 2026
- DocsQuickstart, Helicone documentation · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.