AIInfra（AI 基础设施）指AI系统从底层芯片等硬件，到上层软件栈支持AI大模型训练和推理。

🚀 「大模型」1小时从0训练26M参数的视觉多模态VLM！🌏 Train a 26M-parameter VLM from scratch in just 1 hours!

🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!

深海GLM大模型竞赛

The reinforcement learning training code for AgiBot X1.

The inference module for AgiBot X1.

Pure C++ implementation of several models for real-time chatting on your computer (CPU & GPU)

C++ implementation of Qwen-LM

Tensor library for machine learning

An Autonomous LLM Agent for Complex Task Solving

Langchain-Chatchat（原Langchain-ChatGLM）基于 Langchain 与 ChatGLM, Qwen 与 Llama 等语言模型的 RAG 与 Agent 应用 | Langchain-Chatchat (formerly langchain-ChatGLM), local knowledge based LLM (like ChatGLM, Qwen and…

cube studio开源云原生一站式机器学习/深度学习/大模型AI平台，mlops算法链路全流程，算力租赁平台，notebook在线开发，拖拉拽任务流pipeline编排，多机多卡分布式训练，超参搜索，推理服务VGPU虚拟化，边缘计算，标注平台自动化标注，deepseek等大模型sft微调/奖励模型/强化学习训练，vllm/ollama/mindie大模型多机推理，私有知识库，AI模型市场…

SuperSonic is the next-generation AI+BI platform that unifies Chat BI (powered by LLM) and Headless BI (powered by semantic layer) paradigms.

FlagEval is an evaluation toolkit for AI large foundation models.

基于miniGo的幻影围棋AI，2019中国计算机博弈大赛幻影围棋组冠军；AI of Phantom Go based on miniGo

An open source library for interacting with and processing radar data, specialized for MIMO mmWave radars

hardware and software design of the 250mm autonomous drone

Massively Parallel Deep Reinforcement Learning. 🔥

Making large AI models cheaper, faster and more accessible

Colossal-AI: A Unified Deep Learning System for Big Model Era

搜索所有中文NLP数据集，附常用英文NLP数据集

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

Jittor is a high-performance deep learning framework based on JIT compiling and meta-operators.

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".

OneFlow is a deep learning framework designed to be user-friendly, scalable and efficient.

目标检测数据集制作:VOC,COCO,YOLO等常用数据集格式的制作和互相转换脚本

files_search_engine

MNN is a blazing fast, lightweight deep learning framework, battle-tested by business-critical use cases in Alibaba. Full multimodal LLM Android App:[MNN-LLM-Android](./apps/Android/MnnLlmChat/READ…

机器人视觉 移动机器人 VS-SLAM ORB-SLAM2 深度学习目标检测 yolov3 行为检测 opencv PCL 机器学习 无人驾驶