RAGEN leverages reinforcement learning to train LLM reasoning agents in interactive, stochastic environments.

Environments for LLM Reinforcement Learning

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"

Official code for NeurIPS 2025 paper "GRIT: Teaching MLLMs to Think with Images"

Official Repository of Absolute Zero Reasoner

🔔🧠 Easily experiment with popular language agents across diverse reasoning/decision-making benchmarks!

🌍 Repository for "AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agent", ACL'24 Best Resource Paper.

Agent S: an open agentic framework that uses computers like a human

Open Agent Computer Interface

[NeurIPS 2024] OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments

RL-Agents to play the famous AI-benchmark game hanabi. Contains DQN and PG variants, as well as a Graphical UI, to try out the agents

[ICLR 2024] Source codes for the paper "Building Cooperative Embodied Agents Modularly with Large Language Models"

Colorful Prompt Tuning for Pre-trained Vision-Language Models

Cornell Touchdown natural language navigation and spatial reasoning dataset.

Code for ORAR Agent for Vision and Language Navigation on Touchdown and map2seq

A Probabilistic Programming Language in 70 lines of Python. Code for the blog post https://mrandri19.github.io/2022/01/12/a-PPL-in-70-lines-of-python.html

Localization via embodied dialog on the navigation graph

A music programming language for musicians. 🎶

Train transformer language models with reinforcement learning.

Keyboard using predictive words generated by an NPL model

A minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training