Helicone is an open-source LLM observability platform built around a logging proxy and AI gateway, Apache 2.0 licensed and free to self-host. Routing requests through it captures every model call with cost, latency, and token metrics, adds caching and rate limiting, and provides prompt management on top. The hosted service has a free Hobby tier with a monthly request allowance. Published paid-plan pricing has varied across sources, so current rates should be confirmed with the vendor.
open-sourcellm-tracingllm-gateway
Overview
- Founded
- 2023
- Pricing tier
- Low
- Startup-friendly
- Yes
- Enterprise-ready
- Yes
LLM observability
- Request tracing
- Yes
- Quality evals
- No
- Prompt management
- Yes
- LLM gateway
- Yes
- Open-source
- Yes
- Self-hostable
- Yes
- Free tier
- Yes: Hobby plan: 10k requests/mo; Apache 2.0 self-host free