tensorzero/tensorzero
🏗️ Frameworktensorzero
Open-source LLMOps platform unifying gateway, observability, evaluation, optimization, and experimentation.
TensorZero is an open-source platform designed to streamline LLM operations (LLMOps). It integrates an LLM gateway that routes requests to various model providers (e.g., OpenAI, Anthropic, Llama) while capturing telemetry data for observability. The platform includes evaluation and optimization tools to fine-tune model performance, and experimentation features for A/B testing. Written in Rust for high performance, it offers Python and REST APIs for integration. Key capabilities include prompt management, caching, fallback handling, and real-time feedback collection. TensorZero aims to simplify the lifecycle of LLM applications from development to production.
💡Highlights
- ├─All-in-one LLMOps platform
- ├─Built in Rust for performance
- └─Open-source with 11.6k+ stars
🎯For
- ├─AI engineers
- ├─ML teams
- └─LLM application developers