qwen2-5-vl 开源项目推荐（109 个）

2U1 /Qwen-VL-Series-Finetune

阿里云对Qwen-BL系列进行微调的开源实现。

sophgo /LLM-TPU

Run generative AI models in sophgo BM1684X/BM1688

Brekel /VisionCaptioner

Automated image & video captioning using Qwen-VL, Gemma4 and SAM3.

thaoshibe /relsim

🍑 relsim：关系视觉相似性 |pip install relsim 🌍 （CVPR 2026）

oceanflowlab /OmniVTG

[CVPR 2026] OmniVTG: A Large-Scale Dataset and Training Paradigm for Open-World Video Temporal Grounding

sespoir /ReGround

[ACM MM 2026] ReGround: Restoring Visual Grounding in Multi-Step Reasoning through Self-Diagnosis and Visual Re-Examination

yuanc3 /DATE

Use 2 lines to empower absolute time awareness for Qwen2.5VL's MRoPE

o-l-l-i /simple-captioner

Simple image and video captioning app with a Gradio UI, powered by Qwen2.5/3 VL Instruct.

PRITHIVSAKTHIUR /OCR-ReportLab-Notebooks

A dedicated Colab notebooks to experiment (Nanonets OCR, Monkey OCR, OCRFlux 3B, Typhoo OCR 3B & more..) On T4 GPU - free tier

liuyifan22 /Qwen2.5-VL-Batched

A batched implementation for efficient Qwen2.5-VL inference.

just-a-go /SOVA

Official implementation of SOVA: Strict Outcome-Conditioned Virtual Advantages for Video Reasoning. Correctness-oriented calibration for TW-GRPO without extra rollouts.

PRITHIVSAKTHIUR /Multimodal-OCR

Multimodal-OCR is an experimental, high-performance visual reasoning and optical character recognition suite designed to accurately extract text, analyze visual content, and parse complex document structures. Built upon a diverse ecosystem of cutting-edge vision-language models.

PRITHIVSAKTHIUR /Qwen-Image-Edit-2509-LoRAs-Fast

Qwen-Image-Edit-2509-LoRAs-Fast is a high-performance, user-friendly web application built with Gradio that leverages the advanced Qwen/Qwen-Image-Edit-2509 model from Hugging Face for seamless image editing tasks.

abhinab-choudhury /Crop-AI

Hackathon | SIH 2025 | SIH25030 | "AI-powered crop diagnosis, disease detection, and multilingual farming assistant - empowering farmers with real-time, offline-ready agricultural insights."

zhangguanghao523 /CMMCoT

[AAAI'26] Official implementation of CMMCoT: Enhancing Complex Multi-Image Comprehension via Multi-Modal Chain-of-Thought and Memory Augmentation

cilabuniba /artseek

[WACV 2027] ArtSeek: Deep artwork understanding via multimodal in-context reasoning and late interaction retrieval

PRITHIVSAKTHIUR /Super-OCRs-Demo

基于Gradio的演示应用，用于比较最先进的OCR模型：DeepSeek-OCR、Dots.OCR、HunyuanOCR和Nanonets-OCR2-3B。

PRITHIVSAKTHIUR /QIE-Bbox-Studio

QIE-Bbox-Studio（Qwen 图像编辑边界框 Studio）是一款基于 Qwen2.5-VL 和 Qwen-Image-Edit 模型构建的先进 AI 驱动图像编辑界面。该应用允许用户通过定义边界框和提供自然语言提示，极其精确地操控图像。

PRITHIVSAKTHIUR /Qwen3-VL-Outpost

Qwen3-VL-Outpost is an experimental, high-performance visual reasoning and multimodal inference suite designed for advanced image analysis, optical character recognition, and complex scene understanding. Built around the state-of-the-art Qwen3-VL and Qwen2.5-VL model families.

Kathan-max /RAG-Enhanced-Chatbot-with-LoRA-Fine-Tuning

Transform your documents into intelligent conversations. This open-source RAG chatbot combines semantic search with fine-tuned language models (LLaMA, Qwen2.5VL-3B) to deliver accurate, context-aware responses from your own knowledge base. Join our community!