Directory / Helicone
Helicone
Verified
Open LLM gateway with logging, caching, and spend controls.
Helicone is an open-source observability and gateway layer for LLM APIs: proxy, logs, cache, rate limits. Factories use it as the meter and the flight recorder for inference.
- Model independence
- Router
- Deployment
- Cloud, Self-hosted
- Open source
- Yes
- License
- Apache-2.0
- Pricing model
- Mixed
- Added / updated
- 19 Aug 2026 / 19 Aug 2026
www.helicone.ai · Docs · Source
Editorial take
Helicone is a gateway, which is why it is marked router: it sits in front of many providers. That is observability plus a bit of orchestration. It will not judge whether the patch was correct.
Strengths
- Apache-2.0 proxy you can self-host.
- Cost, latency, and log visibility across providers.
- Caching/rate limits as factory controls.
Limitations
- A proxy outage is an inference outage.
- Not an eval framework by itself.
- Cloud vs OSS.