evalite is a lightweight, model-agnostic framework for evaluating AI agents. It provides developers with tools for defining evaluation tests and measuring agent behavior across models.
evalite is a LLM eval & observability project. It focuses on evaluating AI agent behavior consistently across different models and test cases. It is built as an open-source project for AI developers. The project is open source (MIT). evalite is available on the command line, and it can be self-hosted.
It is developed by githyuvi, and it first shipped in 2026. Development happens publicly on GitHub with 69 commits in the last 90 days. Among its 4 catalogued features are agent evaluation, model agnostic, and evaluation metrics.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do