SpecTrax is a JAX-native library for neural networks and graph learning, built for performance, composability and modularity.

Inference-only port of EasyDeL for Apple Silicon via MLX

Agents for intelligence and coordination

Accelerate, Optimize performance with streamlined training and serving options with JAX.

Cambrian-S: Towards Spatial Supersensing in Video

GEPA extension for pydantic-ai

Second Order Optimization and Curvature Estimation with K-FAC in JAX.

SkyRL: A Modular Full-stack RL Library for LLMs

a Jax quantization library

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

A flexible and efficient implementation of Flash Attention 2.0 for JAX, supporting multiple backends (GPU/TPU/CPU) and platforms (Triton/Pallas/JAX).

TORAX: Tokamak transport simulation in JAX

Ben is nonspeaking and quadriplegic with vision challenges who only has access to left and right head switches (space and return)

llama.cpp fork with additional SOTA quants and improved performance

Sophia optimizer with ScheduleFree

Sharing both practical insights and theoretical knowledge about LLM evaluation that we gathered while managing the Open LLM Leaderboard and designing lighteval!

A comprehensive JAX/NNX library for diffusion and flow matching generative algorithms, featuring DiT (Diffusion Transformer) and its variants as the primary backbone with support for ImageNet train…

A simple formula supports eight types of quantization

A Lightweight LLM Post-Training Library

🐜 Research-friendly Deep Learning framework

lightweight optimizer benchmark