The RL Bridge for LLM-based Agent Applications. Made Simple & Flexible.

[ICLR 2026] Variation in Verification: Understanding Verification Dynamics in Large Language Models

An implementation of the regularization method "AEnt" introduced in "on entropy control in LLM-RL algorithms".

Distributed RL System for LLM Reasoning

The best ChatGPT that $100 can buy.

The best ChatGPT that $100 can buy.

verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework

Search, understand, reproduce, and improve an idea with ease

The official PyTorch implementation of ALRIGHT and MAXRIGHT algorithms for efficient trade-off in LLM post-training

An implementation of SEAL: Safety-Enhanced Aligned LLM fine-tuning via bilevel data selection.

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

Pytorch implementation of AiPOD and E-AiPOD.

An implementation of the penalty-based bilevel gradient descent (PBGD) algorithm and the iterative differentiation (ITD/RHG) methods.

Implemented AdaTD and compared it with other optimization methods in temporal difference learning.