Helicone
Open-source LLM gateway and observability platform for routing, debugging and analyzing AI app requests
Helicone is an LLMOps platform and AI gateway that sits between an app and its model providers. It routes requests across providers such as OpenAI, Anthropic and Azure and logs each one, so developers can see what their LLM calls and agents do in production.
Key features:
- Route requests across model providers with automatic fallbacks.
- Cache responses and apply rate limits at the gateway.
- Inspect requests, sessions and user analytics.
- Query logs with HQL and set up alerts and reports.
- Manage prompts, scores and datasets and test them in a playground.
The scraped homepage is short, so these details come from the brand data and pricing page. Helicone is open source under Apache 2.0, with on-prem deployment on Enterprise. The free Hobby plan includes 10,000 requests a month; Pro is $79/month and Team $799/month with SOC 2 and HIPAA, both plus usage and with a 7-day free trial.