The HELMET Benchmark

A simple, performant and scalable Jax LLM!

A framework for the large scale analysis of programming language usage.

TPU inference for vLLM, with unified JAX and PyTorch support.

C API for MLX

A Lightweight LLM Post-Training Library

Our library for RL environments + evals

A benchmark for LLMs on complicated tasks in the terminal

Implementation of 2-simplicial attention proposed by Clift et al. (2019) and the recent attempt to make practical in Fast and Simplex, Roy et al. (2025)

Nano vLLM

Roo Code gives you a whole dev team of AI agents in your code editor.

Holistic Evaluation of Language Models (HELM) is an open source Python framework created by the Center for Research on Foundation Models (CRFM) at Stanford for holistic, reproducible and transparen…

Minimal yet performant LLM examples in pure JAX

(EasyDel Former) is a utility library designed to simplify and enhance the development in JAX

A lightweight data processing framework built on DuckDB and 3FS.

A toolkit for describing model features and intervening on those features to steer behavior.

AutoBound automatically computes upper and lower bounds on functions.

A framework for few-shot evaluation of language models.

Evaluation suite for LLMs

Implementation of PSGD optimizer in JAX

Lighteval is your all-in-one toolkit for evaluating LLMs across multiple backends

Entropy Based Sampling and Parallel CoT Decoding

Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.