#reinforcement-learning-from-human-feedback Repositories
Discover 33 active open source GitHub repositories tagged with #reinforcement-learning-from-human-feedback, accumulating 86,497 total stars (averaging 2,621 stars per project). Leading this topic is huggingface/trl with 19,788 stars.
33 repositories total
An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Async RL)
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Reference implementation for DPO (Direct Preference Optimization)
HY-Motion model for 3D human motion or 3D character animation generation.
[NeurIPS 2025] An official implementation of Flow-GRPO: Training Flow Matching Models via Online RL
About Awesome things towards foundation agents. Papers / Repos / Blogs / ...
Safe RLHF: Constrained Value Alignment via Safe Reinforcement Learning from Human Feedback
The official repo of Pai-Megatron-Patch for LLM & VLM large scale training developed by Alibaba Cloud.
[ICLR2026] This is the first paper to explore how to effectively use R1-like RL for MLLMs and introduce VisionโR1, a reasoning MLLM that leverages coldโstart initialization and RL training to incentivize reasoning capability.
Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference
ChatLLaMA ๐ข Open source implementation for LLaMA-based ChatGPT runnable in a single GPU. 15x faster training process than ChatGPT
[NeurIPS'23 Spotlight] "Mind2Web: Towards a Generalist Agent for the Web" -- the first LLM-based web agent and benchmark for generalist web agents
Lightweight LLM Post-training Framework, supports SFT, RLVR, On-Policy KD, Guide KD, and mixed training; implements single/multi-round Guide distillation, multi-teacher distillation, Reward mixed training, and automated data sharding๐ฉโ๐๐จโ๐
A simulation framework for RLHF and alternatives. Develop your RLHF method without collecting human data.
Official implementation of UnifiedReward & [NeurIPS 2025] UnifiedReward-Think & UnifiedReward-Flex
This is a collection of research papers for Self-Correcting Large Language Models with Automated Feedback.
Frequently Asked Questions
What is the most popular #reinforcement-learning-from-human-feedback repository on GitHub?
The top ranked open source #reinforcement-learning-from-human-feedback project on GitHub is huggingface/trl with 19,788 stars. Train transformer language models with reinforcement learning.
Which programming languages are most used in #reinforcement-learning-from-human-feedback repositories?
Among the top ranked #reinforcement-learning-from-human-feedback projects on GitHub, the most common programming languages are Python, Jupyter Notebook, HTML, TypeScript.
How is the #reinforcement-learning-from-human-feedback repository ranking calculated?
We track 33 active open source repositories tagged with #reinforcement-learning-from-human-feedback (500+ stars), totaling 86,497 stars with an average of 2,621 stars per repository. Rankings update daily from the GitHub API.