tensorzero
TensorZero is an open-source LLMOps platform that unifies LLM gateway, observability, evaluation, optimization, and experimentation.
- Region
- Overseas
- Pricing
- Free
- Open source
- Yes
- GitHub Stars
- ★ 11.4k
- Source
- GitHub
- Added
- 2026-06-05
- Last verified
- 2026-06-05
Overview
TensorZero is an open-source LLMOps platform integrating LLM gateway, observability, evaluation, optimization, and experimentation. It provides a unified API to access major LLM providers, supporting high-performance calls, tool usage, structured output, and more. Suitable for enterprises and individual developers needing to integrate, monitor, and optimize LLMs.
Key features
- ▪Unified LLM Gateway
- ▪High-performance calls <1ms p99 latency
- ▪Comprehensive observability
- ▪Built-in A/B testing
- ▪Optimize prompts and models
Use cases
Pros
- ＋Supports multiple LLM providers
- ＋Low latency, high performance
- ＋Rich feature modules
- ＋Easy to integrate and extend
Limitations / notes
- －Requires some technical expertise
- －Some advanced features may be complex
Who it's for
This overview was compiled by AI from public sources and may contain inaccuracies — please refer to the official site.
FAQ
Which LLM providers does TensorZero support?
Supports all major LLM providers, including OpenAI SDK and other self-hosted models.
How do I get started with TensorZero?
Visit the official website documentation and follow the quick start guide to configure and deploy.
Something wrong? Let us know on the About page and we'll fix it.