Web based LLM Image Describer - UCAS @2025
Backend Server:
Running
Hugging Face Connection:
Checking...
Select Target Language:
Chinese
English
French
Spanish
German
Select VLM for Image Analysis:
BLIP-Large (Detailed Description)[citation:2]
BLIP2 (VQA & Detailed)[citation:6]
ViT-GPT2 (Fast)
Select Llama Model:
Llama 3.1 8B Instruct[citation:5]
Llama 3.1 70B (Best Quality)
TinyLlama (Fastest)
Image Source:
Start Camera
Stop Camera
Upload Image
Capture Image
Captured Image:
Generate Description
Analysis Results
📷 Basic Image Description:
Generating...
🔍 Detailed Object Analysis:
Generating...
🌍 Translation:
Generating...
📊 JSON Output: