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。
☆875Feb 18, 2025Updated last year
Alternatives and similar repositories for mini_qwen
Users that are interested in mini_qwen are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Train a 1B LLM with 1T tokens from scratch by personal☆816Apr 27, 2025Updated last year
- This is a repository used by individuals to experiment and reproduce the pre-training process of LLM.☆503May 1, 2025Updated last year
- 这是一个open-r1的复现项目，对0.5B、1.5B、3B、7B的qwen模型进行GRPO训练，观察到一些有趣的现象。☆63Apr 13, 2025Updated last year
- 🧠 Train a 64M-parameter LLM from scratch in just 2h!☆62,739Updated this week
- 复现大模型相关算法及一些学习记录☆3,540Jul 2, 2026Updated 2 months ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- 本项目旨在分享大模型相关技术原理以及实战经验（大模型工程化、大模型应用落地）☆25,115Jul 19, 2026Updated 2 months ago
- 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。☆5,843Sep 15, 2026Updated last week
- 从零构建大模型：从预训练到RLHF的完整实践☆2,699Aug 21, 2026Updated last month
- 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☆15,159Jun 14, 2026Updated 3 months ago
- 从零实现一个小参数量中文大语言模型。☆1,089Aug 22, 2024Updated 2 years ago
- 👀 Train a 65M-parameter VLM from scratch in just 2h!☆8,692Updated this week
- Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)☆75,082Sep 14, 2026Updated 2 weeks ago
- 《开源大模型食用指南》针对中国宝宝量身打造的基于Linux环境快速微调（全参数/Lora）、部署国内外开源大模型（LLM）/多模态大模型（MLLM）教程☆32,337Sep 12, 2026Updated 2 weeks ago
- 本项目利用医学领域的 CoT 数据对 Deepseek-R1-Distill-Qwen-7B 进行微调，通过 QLoRA 量化和 Unsloth 加速训练，显著提升模型在复杂医学推理任务中的慢思考能力。知识蒸馏技术使轻量级模型获得大模型的推理优势，实现高效、准确且具有解释性…☆48Mar 10, 2025Updated last year
- 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.
- An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Asy…☆10,045Sep 17, 2026Updated last week
- Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.8, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL…☆15,735Updated this week
- Train deepseek r1-like reasoning LLM with ease | 轻松训练1个deepseek r1类的推理LLM☆21Feb 15, 2025Updated last year
- Qwen3 Fine-tuning: Medical R1 Style Chat☆335May 31, 2025Updated last year
- A comparison of deepseek grpo and qwen gspo on Qwen2.5-1.5B-Instruct fine tunning.☆171Mar 28, 2026Updated 6 months ago
- 《大模型白盒子构建指南》：一个全手搓的Tiny-Universe☆5,077Feb 12, 2026Updated 7 months ago
- verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework☆23,653Updated this week
- 训练一个对中文支持更好的LLaVA模型，并开源训练代码和数据。☆83Sep 6, 2024Updated 2 years ago
- 用于从头预训练+SFT一个小参数量的中文LLaMa2的仓库；24G单卡即可运行得到一个具备简单中文问答能力的chat-llama2.☆2,923Aug 13, 2026Updated last month
- GPUs on demand by Runpod - Special Offer Available • AdRun AI, ML, and HPC workloads on powerful cloud GPUs—without limits or wasted spend. Deploy GPUs in under a minute and pay by the second.
- Reproduce R1 Zero on Logic Puzzle☆2,451Mar 20, 2025Updated last year
- Search-R1: An Efficient, Scalable RL Training Framework for Reasoning & Search Engine Calling interleaved LLM based on veRL☆5,454Nov 13, 2025Updated 10 months ago
- Qwen2.5 0.5B GRPO☆87Feb 16, 2025Updated last year
- 📚 从零开始构建大模型☆34,081Aug 8, 2026Updated last month
- 轻量级 LLM Post-training 框架，支持 SFT、RLVR、On-Policy KD、Guide KD 及混合训练；实现单轮/多轮 Guide 蒸馏、多教师蒸馏、Reward 混合训练与自动化数据分流👩🎓👨🎓☆939Mar 8, 2026Updated 6 months ago
- A travel agent based on Qwen2.5, fine-tuned by SFT + DPO/PPO/GRPO using traveling question-answer dataset, a mindmap can be output using …☆83Jul 6, 2026Updated 2 months ago
- Multi-Modal-AI-Orchestrator (Reset version),AI Full-modal Full-agent:Text → Image → Music → Lights → Video, Includes "Scenario Director,…☆102Nov 5, 2025Updated 10 months ago
- 中文对话0.2B小模型（ChatLM-Chinese-0.2B），开源所有数据集来源、数据清洗、tokenizer训练、模型预训练、SFT指令微调、RLHF优化等流程的全部代码。支持下游任务sft微调，给出三元组信息抽取微调示例。☆1,736Apr 20, 2024Updated 2 years ago
- minimal-cost for training 0.5B R1-Zero☆818May 14, 2025Updated last year
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- 轻量级大语言模型MiniMind的源码解读，包含tokenizer、RoPE、MoE、KV Cache、pretraining、SFT、LoRA、DPO等完整流程☆1,206Jun 16, 2025Updated last year
- 每个人都能看懂的大模型知识分享，LLMs春/秋招大模型面试前必看，让你和面试官侃侃而谈☆7,385Aug 28, 2026Updated last month
- Train a Language Model with GRPO to create a schedule from a list of events and priorities☆272Apr 8, 2026Updated 5 months ago
- 整理开源的中文大语言模型，以规模较小、可私有化部署、训练成本较低的模型为主，包括底座模型，垂直领域微调及应用，数据集与教程等。☆22,771May 10, 2026Updated 4 months ago
- 🏥 从零基础到面试通关：20节课彻底搞懂MedicalGPT医疗大模型训练全流程 | PT/SFT/LoRA/RLHF/DPO/GRPO | 100+面试高频考点☆242Apr 1, 2026Updated 5 months ago
- EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL☆5,171Sep 19, 2026Updated last week
- 将SmolVLM2的视觉头与Qwen3-0.6B模型进行了拼接微调☆612Sep 8, 2025Updated last year