openpsi-project / ReaLHFLinks
Super-Efficient RLHF Training of LLMs with Parameter Reallocation
☆326Updated 7 months ago
Alternatives and similar repositories for ReaLHF
Users that are interested in ReaLHF are comparing it to the libraries listed below
Sorting:
- Implementation for FP8/INT8 Rollout for RL training without performence drop.☆275Updated 3 weeks ago
- A flexible and efficient training framework for large-scale alignment tasks☆440Updated last month
- Bridge Megatron-Core to Hugging Face/Reinforcement Learning☆164Updated last week
- A lightweight reproduction of DeepSeek-R1-Zero with indepth analysis of self-reflection behavior.☆248Updated 7 months ago
- ☆207Updated last month
- A visuailzation tool to make deep understaning and easier debugging for RLHF training.☆271Updated 9 months ago
- Trinity-RFT is a general-purpose, flexible and scalable framework designed for reinforcement fine-tuning (RFT) of large language models (…☆422Updated this week
- Exploring the Limit of Outcome Reward for Learning Mathematical Reasoning☆190Updated 8 months ago
- HuggingFace conversion and training library for Megatron-based models☆228Updated this week
- ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Search (NeurIPS 2024)☆678Updated 10 months ago
- ☆327Updated 6 months ago
- End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning☆330Updated 2 months ago
- Async pipelined version of Verl☆125Updated 7 months ago
- siiRL: Shanghai Innovation Institute RL Framework for Advanced LLMs and Multi-Agent Systems☆279Updated this week
- ☆316Updated 6 months ago
- A version of verl to support diverse tool use☆714Updated last week
- (best/better) practices of megatron on veRL and tuning guide☆103Updated 2 months ago
- Parallel Scaling Law for Language Model — Beyond Parameter and Inference Time Scaling☆456Updated 6 months ago
- Official Repository of "Learning to Reason under Off-Policy Guidance"☆380Updated 2 months ago
- A Comprehensive Survey on Long Context Language Modeling☆209Updated last week
- Reproducing R1 for Code with Reliable Rewards☆275Updated 7 months ago
- ☆213Updated 9 months ago
- MiroRL is an MCP-first reinforcement learning framework for deep research agent.☆180Updated 3 months ago
- L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning☆257Updated 6 months ago
- A lightweight reinforcement learning framework that integrates seamlessly into your codebase, empowering developers to focus on algorithm…☆90Updated 3 months ago
- The related works and background techniques about Openai o1☆221Updated 10 months ago
- Research Code for preprint "Optimizing Test-Time Compute via Meta Reinforcement Finetuning".☆116Updated 4 months ago
- Code for Paper (ReMax: A Simple, Efficient and Effective Reinforcement Learning Method for Aligning Large Language Models)☆198Updated last year
- ByteCheckpoint: An Unified Checkpointing Library for LFMs☆254Updated this week
- ☆65Updated last year