Stars
R-KV: Redundancy-aware KV Cache Compression for Reasoning Models
Brings Git-like branching, versioning, and time-travel capabilities to MongoDB using Docker and S3 for stateless, serverless database environments.
[RA-L 2025 Accept without Revision] A stereo visual-inertial odometry system based on voxel map
[ICML 2025] PyTorch Implementation of "OmniAudio: Generating Spatial Audio from 360-Degree Video"
Train your Agent model via our easy and efficient framework
机场推荐,2025科学上网指南,持续更新,实测高速节点,支持 Clash / V2Ray / Trojan，解锁 ChatGPT / YouTube / Netflix，支持多平台使用，含免费梯子与使用教程。
A lightweight LMM-based Document Parsing Model
AI for Science 论文解读合集（持续更新ing），论文/数据集/教程下载：hyper.ai
zgsm-ai / zgsm
Forked from RooCodeInc/Roo-CodeIntelligent, Practical, and Open-Source Programming Assistant
AppPlatform 是一个前沿的大模型应用工程，旨在通过集成的声明式编程和低代码配置工具，简化和优化大模型的训练与推理应用的开发过程。本工程为软件工程师和产品经理提供一个强大的、可扩展的环境，以支持从概念到部署的全流程 AI 应用开发。
Official Code for Paper《Towards More Faithful Natural Language Explanation Using Multi-Level Contrastive Learning in VQA》
OneFlow is a deep learning framework designed to be user-friendly, scalable and efficient.
Video generation from text&image, 1st-gen
Klavis AI (YC X25): Open Source MCP integration for AI applications
【ICML 2025 Spotlight】 Official Repo for Paper ‘’HealthGPT : A Medical Large Vision-Language Model for Unifying Comprehension and Generation via Heterogeneous Knowledge Adaptation‘’
A Fair and Scalable Time Series Forecasting Benchmark and Toolkit.
面向飞书聊天机器人的全功能AI服务器端实现，用一个容器，实现在飞书对话框里操作属于自己的Manus。
[CVPR 2025 Highlight] Official code for "Olympus: A Universal Task Router for Computer Vision Tasks"
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。