Open Source · Observability
Helicone
by Helicone
An LLM observability platform with an OpenAI-compatible gateway. It logs each request's cost, latency and errors, and routes calls to 100+ models with your own keys or prepaid credits.
Our verdict
An LLM observability platform with an OpenAI-compatible gateway. It logs each request's cost, latency and errors, and routes calls to 100+ models with your own keys or prepaid credits.
Best for: Teams that want request logs, cost tracking and prompt analytics for an LLM app, hosted or self-hosted.
Choose something else if: you need a product with a clear owner and roadmap. Helicone has been in maintenance mode since Mintlify acquired it in March 2026, and since 19 October 2026 its site says it is no longer part of Mintlify, with details to come.
Source review
Using Helicone in practice
Where it runs
Helicone Cloud in US or EU regions, or your own servers. On the hosted gateway, requests pass through Helicone, which logs them along with their cost and latency.
How to install
Hosted: sign up, create a Helicone API key and point your OpenAI client at https://ai-gateway.helicone.ai, then add credits or your own provider keys. Self-hosted: run the all-in-one Docker image and open http://localhost:3000.
Install commands
docker pull helicone/helicone-all-in-one:latest
docker run -d \
--name helicone \
-p 3000:3000 \
-p 8585:8585 \
-p 9080:9080 \
helicone/helicone-all-in-one:latestAs published in the linked sources. Not run by us.
What it costs
Self-hosting is free. Hosted Hobby is free with 10,000 requests and 1 GB of storage. Pro is US$79 a month and Team US$799 a month, plus usage-based charges. Gateway credits carry no token markup, though payment processing fees apply.
Prices and plans change. Check the linked sources before committing.
Why consider it
- Changing one base URL adds logging, cost tracking and alerts to an existing OpenAI-compatible app.
- Hosted in US or EU regions, or self-hosted under the Apache 2.0 licence.
What to watch for
- Maintenance mode means security fixes, bug fixes and new models, not new features. Mintlify said it would help customers move to another platform.
- The self-hosted all-in-one Docker image dates from August 2025, before recent security fixes in the repository, and loses its data on restart unless you mount volumes.
- Prices and fees differ between its pricing page, credits page and docs. Confirm the current terms before you commit.
A first task to try
Point one existing OpenAI call at the Helicone gateway, send a request, then check its cost, latency and log entry in the dashboard.
Suggested exercise, not an installation test. We reviewed the linked documentation; no host compatibility is implied.
Other options to understand
Related tools can solve different parts of the same job. Compare fit before treating them as replacements.
Key facts
What we have checked
- Latest version
- 2025.08.21-1 (released 21 Aug 2025)
- GitHub stars
- 6,177
Each value shows when we last confirmed it and where it came from. Values we have not confirmed are left out, never guessed.
Stars and usage
How Helicone is growing
5,766 to 6,177
Sources
Where this comes from
- GitHubOfficial
- Pricing and ownership noticeOur review
- Hosted quick startOur review
- Self-hosting with DockerOur review
- github api via nightly crawl, recovered from git history 20e874270887c1008f860ee0198e461a23c9e438Registry
A source shows what was published when we checked it. It does not certify security, performance or a working installation. Read our method.
Changelog
What changed
Released 2025.08.21-1
Released 2025.08.21
Released 2025.08.20
Released 1.0.0
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.