Image Chat

Extract information from images using text - an interactive chat interface powered local models.

Quick Start

Setup

Install uv and clone the repository:

git clone <repo-url>
cd image-chat
export PYTORCH_ENABLE_MPS_FALLBACK=1  # For macOS
uv sync

Install Ollama and pull the model:

ollama pull qwen2.5:3b

Run

Launch the ollama in the background.

Terminal 1 - Start MCP server:

uv run python -m detection_mcp_server.main

The server will start on http://127.0.0.1:8000. Verify it's running:

curl http://127.0.0.1:8000/health

Terminal 2 - Start Chat API server:

uv run python -m chat_api.main

Terminal 3 - Start Gradio client:

uv run python -m gradio_chat_client.main

Open http://127.0.0.1:7860 in your browser. Upload an image and chat!

• "What is in this image?"
• "Detect all cars"
• "How many did you find?"