liangyuwang / zo2Links
ZO2 (Zeroth-Order Offloading): Full Parameter Fine-Tuning 175B LLMs with 18GB GPU Memory
☆188Updated 2 months ago
Alternatives and similar repositories for zo2
Users that are interested in zo2 are comparing it to the libraries listed below
Sorting:
- CPPO: Accelerating the Training of Group Relative Policy Optimization-Based Reasoning Models (NeurIPS 2025)☆152Updated 2 weeks ago
- ☆106Updated 3 months ago
- Rethinking RL Scaling for Vision Language Models: A Transparent, From-Scratch Framework and Comprehensive Evaluation Scheme☆142Updated 5 months ago
- ☆202Updated 5 months ago
- The official repo of One RL to See Them All: Visual Triple Unified Reinforcement Learning☆313Updated 4 months ago
- [ICML2025] The official implementation of "C-3PO: Compact Plug-and-Play Proxy Optimization to Achieve Human-like Retrieval-Augmented Gene…☆40Updated 5 months ago
- [ICML 2025] |TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation☆113Updated 4 months ago
- "what, how, where, and how well? a survey on test-time scaling in large language models" repository☆70Updated last week
- A generalized framework for subspace tuning methods in parameter efficient fine-tuning.☆155Updated 3 months ago
- Trinity-RFT is a general-purpose, flexible and scalable framework designed for reinforcement fine-tuning (RFT) of large language models (…☆356Updated last week
- MiroMind-M1 is a fully open-source series of reasoning language models built on Qwen-2.5, focused on advancing mathematical reasoning.☆236Updated last month
- Pre-trained, Scalable, High-performance Reward Models via Policy Discriminative Learning.☆157Updated last week
- Extrapolating RLVR to General Domains without Verifiers☆168Updated last month
- Exploring the Limit of Outcome Reward for Learning Mathematical Reasoning☆190Updated 6 months ago
- MiroRL is an MCP-first reinforcement learning framework for deep research agent.☆162Updated last month
- ☆71Updated 4 months ago
- One-shot Entropy Minimization☆185Updated 3 months ago
- A highly capable 2.4B lightweight LLM using only 1T pre-training data with all details.☆217Updated 2 months ago
- ☆169Updated 5 months ago
- MMSearch-R1 is an end-to-end RL framework that enables LMMs to perform on-demand, multi-turn search with real-world multimodal search too…☆324Updated last month
- DeepSpeed教程 & 示例注释 & 学习笔记 (大模型高效训练)☆178Updated 2 years ago
- An Easy-to-use, Scalable and High-performance RLHF Framework designed for Multimodal Models.☆146Updated 5 months ago
- Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMs☆192Updated 3 months ago
- ☆84Updated 6 months ago
- [COLM 2025] LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation☆149Updated 2 months ago
- MiroTrain is an efficient and algorithm-first framework for post-training large agentic models.☆87Updated last month
- 青稞Talk☆148Updated 2 weeks ago
- MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources☆192Updated last week
- ☆49Updated 2 months ago
- Repo for "VRAG-RL: Empower Vision-Perception-Based RAG for Visually Rich Information Understanding via Iterative Reasoning with Reinforce…☆352Updated 3 months ago