What is Helicone?
LLM usage routed, debugged and analysed.
Who is Helicone best for?
Helicone is best for developers running LLM apps in production who need request logs, cost tracking and a gateway.
Teams shipping LLM features
Every request, response time and cost is logged, so you can trace sessions and users and find slow or failing calls.
Apps on several model providers
The gateway routes calls to OpenAI, Anthropic, Azure, OpenRouter and others, with caching, rate limits and automatic fallbacks.
Early-stage startups
The free plan covers 10,000 requests a month, and young startups can apply for half off the first paid year.
Regulated products
The Team plan adds SOC-2 and HIPAA compliance, and Enterprise can run on your own servers.
How much does Helicone cost?
Helicone's Hobby plan is free with 10,000 requests a month, and its paid plans cost $79 a month for Pro or $799 a month for Team, plus usage-based fees for extra requests and storage.
- 10,000 requests a month
- 1 GB storage, 7-day data retention
- 1 seat, 1 organization
- Unlimited seats, 1-month data retention
- Alerts, reports and HQL queries
- 7-day free trial, usage billed on top
- 5 organizations, 3-month data retention
- SOC-2 and HIPAA compliance
- Dedicated Slack channel, 7-day free trial
Pro and Team include 10,000 requests and 1 GB of storage; after that, requests cost $0.0007 each, falling in tiers to $0.00002 above 2.5 million, and storage costs $3.25 per GB for the first 30 GB, falling to $0.50 per GB above 450 GB. Enterprise is priced on request and adds SAML SSO, on-prem deployment and a custom MSA. Startups under 2 years old with under $5M in funding can get 50% off the first year, students and educators are usually free, and non-profits and open-source companies can ask for discounts or credits. Full pricing on helicone.ai. We checked these prices on .
How do readers rate Helicone?
No one has rated Helicone yet. Be the first.
Sign in to rate. One rating per account.
Is Helicone open source?
Yes. Helicone's code is public under the Apache 2.0 licence, so you can run it on your own server.
On GitHub
6,280stars
- Last commit
- 12 Nov
- Repository age
- 3 years
- Latest release
- 21 Aug 2025
- Licence
- Apache 2.0
- Self-hosted
- Yes
- Repository
- Helicone/helicone
What are the best Helicone alternatives?
Helicone is an open source alternative to Orq.ai, Lunary, Openlayer and Splunk Agent Observability. Langfuse and AgentOps do the same job and are open source too.
Agents built and evaluated with real-time tracing.
Monitoring, analytics and prompt management for agents.
AI governance with evaluations, tracing, guardrails and compliance tracking.
AI agent tracing and evaluations in Splunk Observability Cloud, formerly Galileo.
Open-source tracing, prompt management and evals for LLM apps.
Trace, debug and cost-track your AI agents in production.
Questions about Helicone
What is Helicone?
LLM usage routed, debugged and analysed.
Who is Helicone for?
Helicone is best for developers running LLM apps in production who need request logs, cost tracking and a gateway.
Is Helicone free?
Yes. Helicone's Hobby plan is free with no card, covering 10,000 requests a month, 1 GB of storage, one seat and 7 days of data retention. The 7-day free trial applies to the paid Pro and Team plans.
How much does Helicone cost?
Helicone's paid plans are Pro at $79 per month and Team at $799 per month.
Is Helicone open source?
Yes. Helicone's code is public under the Apache 2.0 licence, so you can run it on your own server.
What are the best Helicone alternatives?
The closest alternatives on Recaply are Orq.ai, Lunary, Openlayer, Splunk Agent Observability, Langfuse and AgentOps.
Where can I try Helicone?
On its official website, helicone.ai.
Is Helicone still maintained after the Mintlify acquisition?
Yes, in maintenance mode. Mintlify bought Helicone and its team moved over, but Helicone says the service stays live and keeps getting security updates, support for new models, and bug and performance fixes. Plans can still be switched or cancelled at any time.





