vwxyzjn

grok-1

Grok open release

llm.c

LLM training in simple, raw C/CUDA

uv

An extremely fast Python package and project manager, written in Rust.

openui

OpenUI let's you describe UI using your imagination, then see it rendered live.

mlx

MLX: An array framework for Apple silicon

mamba

Mamba SSM architecture

minbpe

Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.

cv

Print-friendly, minimalist CV page

PokemonRedExperiments

Playing Pokemon Red with Reinforcement Learning

deskhop

Fast Desktop Switching Device

gpt-fast

Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.

OLMo

Modeling, training, eval, and inference code for OLMo

direct-preference-optimization

Reference implementation for DPO (Direct Preference Optimization)

summarize-from-feedback

Code for "Learning to summarize from human feedback"

huak

My experimental Python package manager.

text-clustering

Easily embed, cluster and semantically label text datasets

aimo-progress-prize

Stoix

🏛️A research-friendly codebase for fast experimentation of single-agent reinforcement learning in JAX • End-to-End JAX RL

flashbax

⚡ Flashbax: Accelerated Replay Buffers in JAX

RL4VLM

Official Repo for Fine-Tuning Large Vision-Language Models as Decision-Making Agents via Reinforcement Learning

fms-fsdp

🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash attention v2.

seqax

seqax = sequence modeling + JAX

oaib

Use the OpenAI Batch tool to make async batch requests to the OpenAI API.

jaxued

sebulba

🪐 The Sebulba architecture to scale reinforcement learning on Cloud TPUs in JAX

paged-attention-minimal

a minimal cache manager for PagedAttention, on top of llama3.

PDFwriter

An OSX print to pdf-file printer driver

cogment-lab

A toolkit for practical Human-AI cooperation research

OLMo-core

PyTorch building blocks for OLMo

putting-dune