Modelpublic

drawais/Qwen3-4B-SpinQuant-NVFP4

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face

0likes25downloads

Model Card

Qwen3-4B-SpinQuant-NVFP4

INT4 weight-only quantization of `Qwen/Qwen3-4B`.

Qwen3-4B in NVFP4 W4A4. Native vLLM compressed-tensors. About 2.5 GB on disk.

Property	Value
Base model	Qwen/Qwen3-4B
Quantization	INT4 weight-only
Approx. on-disk size	~2.8 GB
License	Apache License, Version 2.0
Languages	English

Load (vLLM)

bash

vllm serve drawais/Qwen3-4B-SpinQuant-NVFP4 \
  --max-model-len 32768 \
  --gpu-memory-utilization 0.94

python

from vllm import LLM, SamplingParams
llm = LLM(model="drawais/Qwen3-4B-SpinQuant-NVFP4", max_model_len=32768)
print(llm.generate(["Hello!"], SamplingParams(max_tokens=128))[0].outputs[0].text)

Footprint

~2.8 GB on disk. Recommended VRAM: enough headroom for KV cache.

License & attribution

This artifact is a derivative work of `Qwen/Qwen3-4B`, released by its original authors under the Apache License, Version 2.0.

This artifact is distributed under the same license. The full license text is included in `LICENSE`, and required attribution is in `NOTICE`.

License text: https://www.apache.org/licenses/LICENSE-2.0 Source model: https://huggingface.co/Qwen/Qwen3-4B