Pezzo
Open-source LLMOps platform for prompt management and LLM observability.
About
Pezzo is a fully open-source, cloud-native LLMOps platform that centralizes prompt management, versioning, observability, and caching for teams running large language models in production. It is built in TypeScript and released under the Apache 2.0 license, with the source available at github.com/pezzolabs/pezzo.
The platform provides commit-based prompt versioning with environment-specific publishing, so prompts move from development to staging to production like code. A proxy layer with Redis-backed caching intercepts LLM requests to cut cost and latency, while every request is traced with token counts, cost, and latency into ClickHouse for analytics. It deploys as a Docker Compose stack of NestJS, PostgreSQL, ClickHouse, Redis, and SuperTokens, with Node.js, Python, and LangChain clients.
What it can do
- Version-controlled prompt management with environment publishing
- LLM request observability with token, cost, and latency tracing
- Redis-backed caching proxy to reduce inference cost and latency
- Node.js, Python, and LangChain client SDKs
Setup
Self-host via Docker Compose using the repo quickstart, or run the infrastructure dependencies (PostgreSQL, ClickHouse, Redis, SuperTokens) and the API server separately for development.
Prerequisites
Docker, Node.js 18+, and API keys for your LLM providers.
Use cases
Works with
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 133; Trending score 72/100 from researcher assessment.
★ 3,239 GitHub stars
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.