Helicone acts as a proxy between your application and your LLM provider — change a single API base URL and suddenly every request, response, cost, and latency is captured. The proxy-based architecture means zero code changes (just a config change) and works with any LLM provider that has an OpenAI-compatible API.
Change your OpenAI base URL from api.openai.com to oai.helicone.ai and add your Helicone API key as a header. That's it. Every request now flows through Helicone, which captures metadata and forwards to OpenAI. Custom metrics, properties, and user IDs can be attached via headers. Works with OpenAI, Anthropic, Cohere, and any provider with an OpenAI-compatible endpoint.
Free tier: 100k requests/month. Pro: $20/month (2M requests). Business: $500/month (unlimited + SSO). Self-hosted available.
Teams that want observability without rewriting their code. Particularly popular with early-stage startups and teams that already have instrumented code they don't want to touch.