Package profile
pydantic-evals
- Summary: Framework for evaluating stochastic code execution, especially code making use of LLMs
- Author: Samuel Colvin <samuel@pydantic.dev>, Marcelo Trylesinski <marcelotryle@gmail.com>, David Montague <david@pydantic.dev>, Alex Hall <alex@pydantic.dev>, Douwe Maan <douwe@pydantic.dev>
- License: MIT
- Homepage: https://pydantic.dev/docs/ai/evals/evals/
- Source: https://github.com/pydantic/pydantic-ai
- Number of releases: 294
- First release: 0.0.47 on 2025-03-31
- Latest release: 2.55.0 on 2026-10-09
- Latest release size: 103.4 KB (pure Python wheel)
Dependencies
Pydantic-evals has 7 dependencies, 1 of which optional.Dependent packages
| Package | Optional | Group |
|---|---|---|
| pydantic-ai-examples | false | |
| logfire | true | datasets |
| pydantic-ai-slim | true | evals |
Similar packages
- swebenchThe official SWE-bench package - a benchmark for evaluating LMs on software engineering
- deepevalThe LLM Evaluation Framework
- about-timeEasily measure timing and throughput of code blocks, with beautiful human friendly representations.
- langchain-experimentalBuilding applications with LLMs through composability
- langchainBuilding applications with LLMs through composability
- langchain-coreBuilding applications with LLMs through composability
- langchain-classicBuilding applications with LLMs through composability