YOLO Master WebUI Demo

This Space runs a Gradio-based YOLO Master WebUI demo with an embedded Agent system.

Features

• Single Image Inference — YOLO detect / seg / cls / pose / obb
• Batch Processing — Multi-image batch with summary report
• Video Processing — Frame-by-frame detection with annotated output
• Webcam Capture — Take a photo and run inference
• 🤖 Agent Chat — Natural language interaction with smart parameter recommendation
• 🧠 VLM Agent Report — Visual reasoning report via OpenAI / DashScope / Custom API
• Smart Infer — Agent auto-analyzes image features and tunes conf/iou/max_det
• Auto-Enhance — Adaptive retry based on last inference quality
• History & Logs — Persistent inference history and runtime logs

Configuration

Set the following environment variables (or Hugging Face Spaces Secrets) to enable VLM Agent Reports:

Local Development

pip install -r requirements.txt
python app.py

The app runs on http://localhost:7860 by default.