liangyuwang / zo2Links
ZO2 (Zeroth-Order Offloading): Full Parameter Fine-Tuning 175B LLMs with 18GB GPU Memory
☆171Updated last month
Alternatives and similar repositories for zo2
Users that are interested in zo2 are comparing it to the libraries listed below
Sorting:
- Rethinking RL Scaling for Vision Language Models: A Transparent, From-Scratch Framework and Comprehensive Evaluation Scheme☆138Updated 4 months ago
- ☆199Updated 4 months ago
- ☆103Updated 2 months ago
- CPPO: Accelerating the Training of Group Relative Policy Optimization-Based Reasoning Models☆148Updated 2 months ago
- The official repo of One RL to See Them All: Visual Triple Unified Reinforcement Learning☆310Updated 2 months ago
- "what, how, where, and how well? a survey on test-time scaling in large language models" repository☆62Updated this week
- [ICML 2025] |TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation☆113Updated 3 months ago
- A generalized framework for subspace tuning methods in parameter efficient fine-tuning.☆153Updated last month
- A highly capable 2.4B lightweight LLM using only 1T pre-training data with all details.☆208Updated last month
- This is a repo for showcasing using MCTS with LLMs to solve gsm8k problems☆86Updated 5 months ago
- Pre-trained, Scalable, High-performance Reward Models via Policy Discriminative Learning.☆147Updated last month
- ☆79Updated 4 months ago
- [ICML2025] The official implementation of "C-3PO: Compact Plug-and-Play Proxy Optimization to Achieve Human-like Retrieval-Augmented Gene…☆38Updated 3 months ago
- Official code of *Virgo: A Preliminary Exploration on Reproducing o1-like MLLM*☆105Updated 2 months ago
- [COLM 2025] LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation☆146Updated last month
- ☆67Updated 2 months ago
- Efficient Mixture of Experts for LLM Paper List☆93Updated last week
- MMR1: Advancing the Frontiers of Multimodal Reasoning☆162Updated 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…☆292Updated 2 weeks ago
- Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMs☆188Updated 2 months ago
- MiroMind-M1 is a fully open-source series of reasoning language models built on Qwen-2.5, focused on advancing mathematical reasoning.☆202Updated last week
- ☆173Updated 6 months ago
- qwen-nsa☆74Updated 4 months ago
- 从零到一实现了一个多模态大模型，并命名为Reyes（睿视），R：睿，eyes：眼。Reyes的参数量为8B，视觉编码器使用的是InternViT-300M-448px-V2_5,语言模型侧使用的是Qwen2.5-7B-Instruct，Reyes也通过一个两层MLP投影层连…☆24Updated 6 months ago
- a toolkit on knowledge distillation for large language models☆141Updated this week
- Exploring the Limit of Outcome Reward for Learning Mathematical Reasoning☆188Updated 5 months ago
- Extrapolating RLVR to General Domains without Verifiers☆140Updated last week
- ☆54Updated 11 months ago
- MLLM @ Game☆14Updated 3 months ago
- ☆69Updated 2 months ago