BigHugger
GH Repository · Helicone

helicone

🧊 Open source LLM observability platform. One line of code to monitor, evaluate, and experiment. YC W23 🍓

stars
6,161
30-day movement
+186/day
Related entries
60
Connections
6
node/yarnnodeevalprompt-managementTypeScriptplaygroundopenaiopen-sourcemonitoringlarge-language-modelsprompt-engineeringagent-monitoringanalyticsycombinatorevaluationdockergptllama-indexllm-observabilityllm-costllmlangchainllm-evaluationllmops

Helicone is an open source LLM observability platform that monitors, evaluates, and lets you experiment with LLM usage, advertised as requiring one line of code to integrate. The repository is written in TypeScript with a Node/Yarn toolchain and Docker support, and ships agent-oriented config files like AGENTS.md, CLAUDE.md, cursor rules, and a devcontainer.

You reach for it when you need open source monitoring, cost tracking, and evaluation for LLM and agent applications.

Use it to

  • Monitor LLM API usage and costs
  • Evaluate model and prompt performance
  • Manage and iterate on prompts
  • Track agent behavior in production
  • Experiment in an LLM playground

For Developers building LLM applications and agents

Role
eval
Language
TypeScript
Licence
Apache-2.0
Forks
670
Open issues
57
Last push
2026-09-16
Latest release
v1.0.0 · 2025-08-19
topicsllm-observabilitymonitoringevaluationprompt-managementanalyticsopen-source