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

simulator for low precision floats and corresponding BLAS

Efficient optimizers

SGLang is a fast serving framework for large language models and vision language models.

Python SDK for building and deploying serverless applications on the Targon platform.

Differentiable ODE solvers with full GPU support and O(1)-memory backpropagation.

⚡️Optimizing einsum functions in NumPy, Tensorflow, Dask, and more with contraction order optimization.

easiest way to calculate gns in pytorch

Code accompanying the paper "Generalized Interpolating Discrete Diffusion"

Multi-Threaded FP32 Matrix Multiplication on x86 CPUs

an old idea from 2022 -- we have LRA we have kron -- just use a LRA per kron factor

minibatch ode with block causal transformers :00:

A collection of niche / personally useful PyTorch optimizers with modified code.

Fast, Modern, and Low Precision PyTorch Optimizers

Deep Networks Grok All the Time and Here is Why

PageRank for LLMs

Mixbox is a library for natural color mixing based on real pigments.

Puffing up reinforcement learning

A collection of hash and RNG functions that rely on hardware AES

supporting pytorch FSDP for optimizers

unofficial implementation of MARS-AdamW in PyTorch

realtime latent world model inference demo

An implementation of PSGD Kron second-order optimizer for PyTorch