Multimodal Chat (SmolVLM)
ReadyUpload images and chat multi-turn with SmolVLM-256M: charts, handwriting, scenes and objects.
This demo loads a large model in your browser — the first run may take a while
Model size ~0.5 GB, please be patient while it loads
WebGPU not available — falling back to CPU (slower)
WASM（较慢） SmolVLM-256M (~500MB), supports multi-image, multi-turn conversations.
WebGPU is not supported in this browser; falling back to WASM (SmolVLM will be very slow, Chrome/Edge recommended).
The first run downloads SmolVLM-256M (~500MB).
Images
Upload images and type a question to start chatting (load a model first)