VLM Browser Lab
Home
Browser-LLM
Realtime OCR
AI Text detection
VLM Browser Lab
5 models Β· sequential Β· client-side only Β· no server needed
π·
Take a photo or upload an image
tap to open camera Β· jpg / png / webp
βΆ Run all models
β Clear
prompt:
backend:
WebGPU
WASM (CPU)
β
0/4
select models to run
SmolVLM 256M
SmolVLM 500M
SmolVLM2 500M
Florence-2 Base
fastest:
β
avg:
β
done:
β