The world's first agentic reverse engineer.

An interface library for RL post training with environments.

Agentic RL Training at Scale

Our library for RL environments + evals

Reproducible, flexible LLM evaluations

Official repository for DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research

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

verl-agent is an extension of veRL, designed for training LLM/VLM agents via RL. verl-agent is also the official code for paper "Group-in-Group Policy Optimization for LLM Agent Training"

PyTorch-native post-training at scale

Learn the building blocks of how to build DeepSeek from scratch.

Fast C++ Pytorch extension for differentiable synthetic aperture radar image formation and autofocus library on CPU and GPU

The Core Flight System (cFS)

Preempt-RT Kernel Build Guide for NVIDIA Development Board

Building Open LLM Web Agents with Self-Evolving Online Curriculum RL

NeXT hardware emulator for a NeXT Cube and NeXT Station. Mirrored from SourceForge

AgentLab: An open-source framework for developing, testing, and benchmarking web agents on diverse tasks, designed for scalability and reproducibility.

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Align Anything: Training All-modality Model with Feedback

Train transformer language models with reinforcement learning.

An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Async RL)

🌎💪 BrowserGym, a Gym environment for web task automation

A project that provides help for using DeepMind's mctx on gym-style environments.

[NeurIPS 2022] 🛒WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents

Recources to build the MFOS - Noise Toaster Synth by Ray Wilson

(ICML 2024) Alphazero-like Tree-Search can guide large language model decoding and training

A library for advanced large language model reasoning

An extensible benchmark for evaluating large language models on planning

[NeurIPS 2023 Spotlight] LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarios (awesome MCTS)

Dopamine is a research framework for fast prototyping of reinforcement learning algorithms.