Helicone is an open source LLM observability platform that provides an API gateway, logging, and analytics with one-line integration for multiple providers. It supports self-hosting via Docker or Helm and offers integration breadth across providers and frameworks.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
Helicone is an AI Gateway & LLM Observability Platform for AI Engineers. It provides a unified API for 100+ models through an AI gateway with intelligent routing and automatic fallbacks, plus observability for traces, sessions, and cost/latency metrics. The project includes features for prompt management, fine-tuning integration, and an enterprise-ready posture (SOC 2 and GDPR compliant).
How it works
Helicone enables logging of requests from various LLM providers through a single integration. It highlights:
- Quick integration: One-line of code to log all your requests from multiple providers and frameworks.
- Observability: Inspect and debug traces and sessions for agents, chatbots, document processing pipelines, and more.
- Analyze: Track cost, latency, quality, and other metrics, with an option to export to PostHog for dashboards.
The repository architecture identifies five services: Web (NextJS), Worker (Cloudflare Workers), Jawn (log collection server), Supabase (auth/db), ClickHouse (analytics), and Minio (object storage).
Getting started
Getting started steps from the README include:
- Get your API key by signing up and add credits.
- Update the baseURL in your code and add your API key.
Code example provided:
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://ai-gateway.helicone.ai",
apiKey: process.env.HELICONE_API_KEY,
});
const response = await client.chat.completions.create({
model: "gpt-4o-mini",
messages: [{ role: "user", content: "Hello!" }]
});
- View logs at Helicone dashboard and access 100+ models through one API.
For self-hosting, Helicone supports:
- Docker: Docker Compose setup with steps to clone, copy env file, and start services via helicone-compose.sh.
- Helm: Production-ready Helm chart available (enterprise@helicone.ai for access).
Docker-related commands from the README include:
# Clone the repository
git clone https://github.com/Helicone/helicone.git
cd docker
cp .env.example .env
# Start the services
./helicone-compose.sh helicone up
Recent releases
Latest releases listed:
- v2025.08.21-1 (2025-08-21)
- v2025.08.21 (2025-08-21)
- v2025.08.20 (2025-08-20)
- v1.0.0 Docker image release (2025-08-19)
Traction
GitHub stars: 6033. (Star count provided in facts.)
Behind the repo
The repo is affiliated with Y Combinator (badge shown), indicating YC backing. It supports integrations across multiple providers and frameworks as listed in the Integrations sections.
Caveats
License: Apache-2.0. Created in 2023-01-31. Last push: 2026-07-25. Open issues: 158. Language: TypeScript. Docker/Helm self-host options exist; manual deployment is not recommended. The README lists extensive integrations and deployment architectures, including five services: Web, Worker, Jawn, Supabase, ClickHouse, Minio.