Helicone
An open-source AI gateway and LLM observability platform that logs, routes, and analyzes requests across model providers with a single line of code.
Helicone's gateway gives access to more than 100 AI models through one API key and the OpenAI API format, with intelligent routing and automatic fallbacks between providers, while its observability side logs requests from OpenAI, Anthropic, LangChain, Gemini, and the Vercel AI SDK with a one-line integration. It inspects and debugs traces and sessions for agents, chatbots, and document pipelines, tracks cost, latency, and quality metrics with one-line export to PostHog for custom dashboards, and includes a playground for iterating on prompts, sessions, and traces.
Its prompt management feature versions prompts using production data and deploys them through the gateway without a code change, and it supports fine-tuning through partners including OpenPipe. It can be self-hosted with a bundled Docker Compose file, or run with a Helm chart for enterprise workloads, and is SOC 2 and GDPR compliant.
It is written in TypeScript and distributed under the Apache-2.0 license, with a free monthly tier of 10,000 requests.