Custom GPTQ-int4 quant (group 32, byte-sliced nibbles, single tied q4 copy of
the 248k-vocab embedding serving both the gather and the logits cascade) of
Qwen/Qwen3.5-2B (text part), in the
wire format of the
qwen35-2b-webgpu browser runtime.
Try it:
https://huggingface.co/spaces/borkiss/qwen35-2b-webgpu-demo · run /web/bench.html
on your GPU and share the .log.
