Stars
TVM Documentation in Chinese Simplified / TVM 中文文档
Matrix-Game: Interactive World Foundation Model
Klavis AI (YC X25): Open Source MCP integration for AI applications
Official Code for “Pixel to Gaussian: Ultra-Fast Continuous Super-Resolution with 2D Gaussian Modeling”
AI-powered tool for efficient abstract and PDF screening in systematic reviews.
FIT: 企业级AI开发框架，提供多语言函数引擎（FIT）、流式编排引擎（WaterFlow）及Java生态的LangChain替代方案（FEL）。原生/Spring双模运行，支持插件热插拔与智能聚散部署，无缝统一大模型与业务系统。
Officially implement of the paper "DriVerse: Navigation World Model for Driving Simulation via Multimodal Trajectory Prompting and Motion Alignment"
[CVPR 2025] Official implementation for "Empowering LLMs to Understand and Generate Complex Vector Graphics" https://arxiv.org/abs/2412.11102
Skywork-R1V2:Multimodal Hybrid Reinforcement Learning for Reasoning
[ECCV 2024 Oral] MotionDirector: Motion Customization of Text-to-Video Diffusion Models.
[CVPR 2025 Highlight] Official code for "Olympus: A Universal Task Router for Computer Vision Tasks"
[NeurIPS 2024 Best Paper][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ult…
[CVPR 2025 Best Paper Award Candidate] VGGT: Visual Geometry Grounded Transformer
OneFlow is a deep learning framework designed to be user-friendly, scalable and efficient.
HeFlwr: Federated Learning for Heterogeneous Devices
🔥🔥🔥 Free open source and easy-to-use laravel eCommerce platform, Base on the Laravel . It supports multiple languages and currencies, integrates ChatGPT OpenAI. The platform features customizable v…
First 3D AI Agent Platform, allowing users to recreate and display Bynce Agent. Using 3D Motion Capture & MMD Model Technology to interact with Bynce.
Mulberry, an o1-like Reasoning and Reflection MLLM Implemented via Collective MCTS
Build multimodal language agents for fast prototype and production
Qihoo360 / 360-LLaMA-Factory
Forked from hiyouga/LLaMA-Factoryadds Sequence Parallelism into LLaMA-Factory
工作流引擎对内提供单位/机关流程管理规则和内部业务流程的数字化落地实践；对外提供自动化地第三方业务驱动、接口接入和算法单元驱动能力。工作流引擎在提供底层驱动引擎的同时对全局透明监控、安全防御和国产化特色功能进行充分考虑，是内部流程管理和业务算法驱动的不二之选。
Align Anything: Training All-modality Model with Feedback
FAST-LIVO2: Fast, Direct LiDAR-Inertial-Visual Odometry
A powerful baseline for image classification, face recognition and image retrieval with Pytorch