Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.

Bagz is a format for storing a sequence of string records. It supports per-record compression and fast index-based lookup.

Quine's fork of "Hyper" by Vercel, A terminal built on web technologies

Builder and index for PyTorch packages

General Agents Python API client

better multiprocessing and multithreading in Python

Python monorepo template with uv

🐶 Automated code review tool integrated with any code analysis tools regardless of programming language

An extremely fast Python linter and code formatter, written in Rust.

🚀 PR Agent - The Original Open-Source PR Reviewer. This repo is not the Qodo free tier! Try the free version on our website.

Example of a monorepo setup for python projects using uv

A simple, performant and scalable Jax LLM!

Pioneering Automated GUI Interaction with Native Agents

An implementation of PSGD Kron in JAX for distributed training in JAX or Flax

Collection of eclectic utils for python.

Large language models (LLMs) made easy, EasyLM is a one stop solution for pre-training, finetuning, evaluating and serving LLMs in JAX/Flax.

Distributed Task Queue (development branch)

Simple job queues for Python

gRPC to JSON proxy generator following the gRPC HTTP spec

seqax = sequence modeling + JAX

Building blocks for productive research

Fast dataset format and loader

Official codebase used to develop Vision Transformer, SigLIP, MLP-Mixer, LiT and more.

ClearML - Auto-Magical CI/CD to streamline your AI workload. Experiment Management, Data Management, Pipeline, Orchestration, Scheduling & Serving in one MLOps/LLMOps solution

Inference Llama 2 in one file of pure 🔥

Open weights language model from Google DeepMind, based on Griffin.

PostgreSQL extension for vscode providing explorer, highlighting, diagnostics, and intellisense

jax-triton contains integrations between JAX and OpenAI Triton

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

Fast and reliable distributed systems in Python