infrastructure

Helicone

by Helicone

Helicone is an open-source LLM observability platform and AI gateway/reverse proxy that intercepts API requests to LLM providers — capturing logs, costs, and latency with under 1ms overhead in self-hosted mode. It integrates via a simple base URL swap (zero code changes), supports 12+ providers including OpenAI and Anthropic, and adds capabilities such as prompt caching, rate limiting, user tracking, and fine-tuning dataset creation. The gateway layer is written in Rust for minimal footprint (~64MB memory) and is available as a cloud-hosted SaaS or fully self-hostable.

8 Overall Score

Scores

Capability
7
Ease of Use
9
Documentation
8
Reliability
8
Value
9
Momentum
8

Details

Status
active
Pricing
freemium
Launch Date
2022-11
Last Updated
2026-03-15

Key Features

  • Zero-code integration via base URL swap supporting 12+ LLM providers
  • Request logging with cost, latency, and token tracking per call
  • Prompt and response caching to reduce costs and latency
  • Rate limiting, user tracking, and custom metadata tagging
  • Rust-powered gateway with sub-1ms overhead and ~64MB memory footprint; fully self-hostable

Tech Stack

RustTypeScriptNext.jsPostgreSQLClickHouseNode.js