Runs in the browser · Vision + language
Run Qwen3.5-0.8B in your browser
Qwen3.5-0.8B downloads 616.9 MB over WebGPU (q4f16), or 1.20 GB over WebAssembly (q8), to read an image and answer questions about it in text entirely in the tab with Transformers.js. No install, no server.
Vision + language · onnx-community/Qwen3.5-0.8B-ONNX
- Parameters
- 873.44M
- Pipeline task
- custom
Reading
Vision-language chat has no stable transformers.js pipeline() task yet; use the model's AutoModel classes directly.
The largest vision + language download in the catalog (616.9 MB).
The larger build here clears 1 GB (1.20 GB); browser cache and memory limits make this a real download to budget for, not an instant load.
The quant ladder spans 5.3x: 616.9 MB (q4f16) to 3.18 GB (fp32).
Will it run in your browser?
Checking for WebGPU support…
All measured variants
| Variant | Size |
|---|---|
| q4f16 WebGPU pick | 616.9 MB |
| q4 | 684.4 MB |
| q8 WASM pick | 1.20 GB |
| fp16 | 2.07 GB |
| fp32 | 3.18 GB |
Sizes measured from the HuggingFace API file tree for onnx-community/Qwen3.5-0.8B-ONNX, not estimated.
Use it with Transformers.js
There is no stable transformers.js pipeline() call for this model yet, and no verified low-level snippet either (see the Reading panel above). Check the model card on HuggingFace for current usage examples.
Sources
Same job, different size
FAQ
WebGPU or WASM for Qwen3.5-0.8B, in practice?
Chrome, Edge, and Safari 26+ run the WebGPU build (q4f16) on the GPU. Firefox has WebGPU on Windows and Apple Silicon Macs but not everywhere yet. Any browser without a WebGPU adapter falls back automatically to the WebAssembly build (q8) on CPU: same model, slower to load, slower to run.
What does q4f16 mean for Qwen3.5-0.8B?
The WebGPU build here uses q4f16: 4-bit weights with some layers kept at 16-bit for stability: WebGPU's usual smallest clean build. The WASM fallback uses q8: 8-bit integers: about a quarter the size of fp32, with a smaller quality trade than 4-bit.
Sizes measured 2026-08-01 from the HuggingFace API. Last validated 2026-08-03. See all browser models or the methodology.