A comparison of deepseek grpo and qwen gspo on Qwen2.5-1.5B-Instruct fine tunning.
☆170Mar 28, 2026Updated 5 months ago
Alternatives and similar repositories for grpo_reproduce
Users that are interested in grpo_reproduce are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ☆41Nov 16, 2025Updated 9 months ago
- Pretrain、Posttrain、RAG、Agent等大模型相关的基础项目合集☆39Dec 7, 2025Updated 8 months ago
- ☆18Mar 15, 2026Updated 5 months ago
- Local-first interview recording review reports with a Codex skill and CLI.☆79May 16, 2026Updated 3 months ago
- Fine-tune Qwen2.5-VL-7B on custom visual QA tasks using LoRA + Accelerate, supporting single/multi-GPU training on COCO 2014 dataset.☆30Apr 28, 2025Updated last year
- End-to-end encrypted cloud storage - Proton Drive • AdSpecial offer: 40% Off Yearly / 80% Off First Month. Protect your most important files, photos, and documents from prying eyes.
- MedicalGPT: Training Your Own Medical GPT Model with ChatGPT Training Pipeline. 训练医疗大模型，实现了包括增量预训练(PT)、有监督微调(SFT)、RLHF、DPO、ORPO、GRPO。☆5,760Jun 3, 2026Updated 2 months ago
- some notes for opensource llm technical reorts☆20Mar 6, 2026Updated 5 months ago
- Personal Project: MPP-Qwen14B & MPP-Qwen-Next(Multimodal Pipeline Parallel based on Qwen-LM). Support [video/image/multi-image] {sft/conv…☆686Mar 10, 2025Updated last year
- A very simple GRPO implement for reproducing r1-like LLM thinking.☆1,703Nov 21, 2025Updated 9 months ago
- 基于Qwen2+SFT+DPO的医疗问答系统，项目中使用了自定义的 SFTTrainer/DPOTrainer/TRPOTrainer用于训练，其次，项目还调用各种知识库工具（neo4j, milvus, LDA, 等）进行自动化训练数据生成。另外，使用 vllm 用于推理…☆88Apr 29, 2026Updated 4 months ago
- 科大讯飞多模态RAG图文问答挑战赛☆75Aug 4, 2025Updated last year
- 复现大模型相关算法及一些学习记录☆3,509Jul 2, 2026Updated last month
- 这是一个从零开始构建的强化学习人类反馈（RLHF）学习代码库，实现了 PPO、GRPO、GSPO 以及相关的策略优化算法，并 提供了清晰、可复现的训练流程。由于文档是由latex文件转译过来，如果md文件渲染异常，请用VScode的md插件打开☆89Dec 19, 2025Updated 8 months ago
- WebResearcher: An Iterative Deep-Research Agent，迭代式深度研究智能体☆50Feb 13, 2026Updated 6 months ago
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- 超简单使用监督微调SFT和强化学习RL去训练领域Agent☆34Oct 20, 2025Updated 10 months ago
- Multi-Modal-AI-Orchestrator (Reset version),AI Full-modal Full-agent:Text → Image → Music → Lights → Video, Includes "Scenario Director,…☆103Nov 5, 2025Updated 9 months ago
- 大学Latex答辩模版 ，当前包含川大、哈工大、中科大。☆11Jul 22, 2024Updated 2 years ago
- 这是一个从头训练大语言模型的项目，包括预训练、微调和直接偏好优化，模型拥有1B参数，支持中英文。☆867Feb 18, 2025Updated last year
- Codex skill for reconstructing diagram images into editable Draw.io files☆24Updated this week
- Qwen2.5 0.5B GRPO☆86Feb 16, 2025Updated last year
- This project extracts feature points based on the Mediapipe open-source model, uses LSTM, CNN, and Vision of Transformer network models f…☆17Oct 11, 2025Updated 10 months ago
- 2024百度商业AI技术创新大赛赛道一：基于大模型的广告检索全国一等奖获奖方案☆19Feb 23, 2025Updated last year
- My Solution and Notes for the Stanford CS336: LLM from scratch☆266Mar 23, 2026Updated 5 months ago
- Deploy open-source AI quickly and easily - Special Bonus Offer • AdRunpod Hub is built for open source. One-click deployment and autoscaling endpoints without provisioning your own infrastructure.
- official implementation of "CLIP-VQDiffusion : Langauge Free Training of Text To Image generation using CLIP and vector quantized diffusi…☆19Sep 5, 2024Updated last year
- Implementing DeepSeek R1's GRPO algorithm from scratch☆1,897Apr 18, 2025Updated last year
- 基于YOLOv5的交通道路目标检测和数据分析软件☆19Feb 27, 2025Updated last year
- 将SmolVLM2的视觉头与Qwen3-0.6B模型进行了拼接微调☆608Sep 8, 2025Updated 11 months ago
- ☆22Dec 18, 2025Updated 8 months ago
- 🧠 Train a 64M-parameter LLM from scratch in just 2h!☆55,356Updated this week
- MRCPSP: This is an implementation of multi-mode resource constrained project scheduling problem (MRCPSP) in MATLAB.☆12May 10, 2019Updated 7 years ago
- 简单易理解的代码，用于在qwen上使用grpo加强数学能力☆60May 14, 2025Updated last year
- Optimization Case Studies: Generic Time Scheduling Problem (GTSP), Resource-Constrained Project Scheduling Problem (RCPSP) with Pulse Var…☆11Nov 7, 2018Updated 7 years ago
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- 基于LangGraph开发的智能体项目，可借助大模型自动调用工具规划旅游行程，包括景点搜索、交通查询、饭店酒店查询等功能☆43Aug 27, 2024Updated 2 years ago
- 主要记录大语言大模型（LLMs） 算法（应用）工程师相关的知识及面试题☆14,980Jun 14, 2026Updated 2 months ago
- ☆108Jul 24, 2025Updated last year
- 包含了LLM的一些手撕代码，如强化学习。可以帮助从代码层面深入理解原理，以及有助于准备大模型面试可能出现的手撕。后续会更新Transformer等更多手撕☆129Mar 15, 2026Updated 5 months ago
- 集成Qwen与DeepSeek等先进大语言模型，支持纯LLM+分类层模式及LLM+LoRA+分类层模式，使用transformers模块化设计和训练便于根据需要调整或替换组件。☆21Sep 1, 2025Updated 11 months ago
- 基于轻量级 Qwen2.5-0.5B 和 SigLIP 的视觉语言多模态模型实现,包含训练和 SFT 代码。分享训练和 SFT 相关代码,记录一下探索和学习的过程。欢迎一起交流讨论~☆22Aug 31, 2025Updated last year
- 🚀 2026届大模型算法岗实习面经 | 包含 DeepSeek/Qwen 技术报告解析、手撕 PPO/RoPE/Transformer、RLHF 核心与八股文 | 持续更新中...☆659Mar 28, 2026Updated 5 months ago