kyaky
/

Qwen3.6-35B-A3B-NVFP4

Text Generation

compressed-tensors

Mixture of Experts

8-bit precision

Model card Files Files and versions

Qwen3.6-35B-A3B-NVFP4 / benchmark.png

kyaky's picture

NVFP4 self-quant (llm-compressor): FP8 attn/GDN + NVFP4-W4A16 experts; beats redhat/unsloth on quality+speed+size

894cdfa verified 4 days ago

History Contribute Delete

54.5 kB