GitHub Open Source Insight
Computed at Sep 11, 2026GitHub repo leaderboard by stars, growth rate and activity.
haotian-liu
[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.
Python25,018 starslast pushed 2 years agoOpenGVLab
[CVPR 2024 Oral] InternVL Family: A Pioneering Open-Source Alternative to GPT-4o. 接近GPT-4o表现的开源多模态对话模型
Python10,156 starslast pushed 12 months agojingyaogong
👀 Train a 65M-parameter VLM from scratch in just 2h!
Python8,608 starslast pushed 1 month agovolcengine
MineContext is your proactive context-aware AI partner（Context-Engineering+ChatGPT Pulse）
Python5,506 starslast pushed 4 months agooff-grid-ai
The Swiss Army Knife of Offline AI. Chat, see, speak, and generate images on your phone or Mac — GGUF LLMs, vision, Whisper speech-to-text, Stable Diffusion, tool calling, and local-network servers. Runs on your CPU, GPU, or NPU. No account, no API key, zero data leaves your device.
TypeScript3,078 starslast pushed 3 hours agoInternLM
InternLM-XComposer2.5-OmniLive: A Comprehensive Multimodal System for Long-term Streaming Video and Audio Interactions
Python2,927 starslast pushed 1 year agoBAAI-Agents
The Cradle framework is a first attempt at General Computer Control (GCC). Cradle supports agents to ace any computer task by enabling strong reasoning abilities, self-improvment, and skill curation, in a standardized general environment with minimal requirements.
Python2,581 starslast pushed 2 years agoNVIDIA-AI-Blueprints
NVIDIA AI Blueprint for video search and summarization (VSS) is a GPU-accelerated reference architecture for building video analytics agents with real-time verified alerts, visual Q&A, and automated reporting. The VSS Blueprint uses vision language models (VLMs) such as NVIDIA Cosmos, LLMs such as NVIDIA Nemotron, RAG, and NVIDIA NIMs.
Python1,855 starslast pushed 3 hours agoAnionex
[dsh]为纯文本模型设计更强大的视觉工具箱：一行安装使用、粘贴图片直接识别、多张图片问答、截图到前端UI 还原等｜DeepSeek Harness-native integration for agent-vision-toolkit: image Q&A, long-screenshot OCR, UI restoration, grounding, pixel diff, Artifacts, and Web UI.
TypeScript872 starslast pushed 20 hours agoMirroS-Lab
S-Space: Exploring Spatial Workspace in Multimodal Models
Python170 starslast pushed 3 days ago
| # | Repo | Language | Stars | 30-day trend | Last updated |
|---|---|---|---|---|---|
| 1 | haotian-liu [NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond. | Python | 25,018 | last pushed 2 years ago | |
| 2 | OpenGVLab [CVPR 2024 Oral] InternVL Family: A Pioneering Open-Source Alternative to GPT-4o. 接近GPT-4o表现的开源多模态对话模型 | Python | 10,156 | last pushed 12 months ago | |
| 3 | jingyaogong 👀 Train a 65M-parameter VLM from scratch in just 2h! | Python | 8,608 | last pushed 1 month ago | |
| 4 | volcengine MineContext is your proactive context-aware AI partner（Context-Engineering+ChatGPT Pulse） | Python | 5,506 | last pushed 4 months ago | |
| 5 | off-grid-ai The Swiss Army Knife of Offline AI. Chat, see, speak, and generate images on your phone or Mac — GGUF LLMs, vision, Whisper speech-to-text, Stable Diffusion, tool calling, and local-network servers. Runs on your CPU, GPU, or NPU. No account, no API key, zero data leaves your device. | TypeScript | 3,078 | last pushed 3 hours ago | |
| 6 | InternLM InternLM-XComposer2.5-OmniLive: A Comprehensive Multimodal System for Long-term Streaming Video and Audio Interactions | Python | 2,927 | last pushed 1 year ago | |
| 7 | BAAI-Agents The Cradle framework is a first attempt at General Computer Control (GCC). Cradle supports agents to ace any computer task by enabling strong reasoning abilities, self-improvment, and skill curation, in a standardized general environment with minimal requirements. | Python | 2,581 | last pushed 2 years ago | |
| 8 | NVIDIA-AI-Blueprints NVIDIA AI Blueprint for video search and summarization (VSS) is a GPU-accelerated reference architecture for building video analytics agents with real-time verified alerts, visual Q&A, and automated reporting. The VSS Blueprint uses vision language models (VLMs) such as NVIDIA Cosmos, LLMs such as NVIDIA Nemotron, RAG, and NVIDIA NIMs. | Python | 1,855 | last pushed 3 hours ago | |
| 9 | Anionex [dsh]为纯文本模型设计更强大的视觉工具箱：一行安装使用、粘贴图片直接识别、多张图片问答、截图到前端UI 还原等｜DeepSeek Harness-native integration for agent-vision-toolkit: image Q&A, long-screenshot OCR, UI restoration, grounding, pixel diff, Artifacts, and Web UI. | TypeScript | 872 | last pushed 20 hours ago | |
| 10 | MirroS-Lab S-Space: Exploring Spatial Workspace in Multimodal Models | Python | 170 | last pushed 3 days ago |
All · 11,658