A Gym for Agentic LLMs

A JAX-native LLM Post-Training Library

Asyncer, async and await, focused on developer experience.

Typer, build great CLIs. Easy to code. Based on Python type hints.

Official implementation of Regularized Policy Gradient (RPG) (https://arxiv.org/abs/2505.17508)

Quickly rewrite git repository history (filter-branch replacement)

Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

The Postgres development platform. Supabase gives you a dedicated Postgres database to build your web, mobile, and AI applications.

Lightweight coding agent that runs in your terminal

DataFlex is a data-centric training framework that enhances model performance by either selecting the most influential samples, optimizing their weights, or adjusting their mixing ratios.

Code for the paper "ClinicalBench: Can LLMs Beat Traditional ML Models in Clinical Prediction?"

Atropos is a Language Model Reinforcement Learning Environments framework for collecting and evaluating LLM trajectories through diverse environments

Checkpoint-engine is a simple middleware to update model weights in LLM inference engines

[ICML 2024] LESS: Selecting Influential Data for Targeted Instruction Tuning

Easy Data Preparation with latest LLMs-based Operators and Pipelines.

Trae Agent is an LLM-based agent for general purpose software engineering tasks.

VeOmni: Scaling Any Modality Model Training with Model-Centric Distributed Recipe Zoo

LLM inference in C/C++

[Preprint] On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification.

Renderer for the harmony response format to be used with gpt-oss

Collection of scripts and notebooks for OpenAI's latest GPT OSS models

gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

The official code implementation of the ACL2025 paper “A Text is Worth Several Tokens: Text Embedding from LLMs Secretly Aligns Well with The Key Tokens”. Text Embedding from LLMs Secretly Aligns W…

A lightweight, local-first, and free experiment tracking library from Hugging Face 🤗

✨ Agentic Reinforced Policy Optimization

An open-source AI agent that brings the power of Gemini directly into your terminal.

Qwen Code is a coding agent that lives in the digital world.

Text-audio foundation model from Boson AI

一个面向多模态大模型训练的智能数据集构建与评估平台