Instructions to use intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound")
model = AutoModelForMultimodalLM.from_pretrained("intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound

SGLang

How to use intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound with Docker Model Runner:
```
docker model run hf.co/intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound
```

Model Details

This model is an NVFP4+ (also called uNVFP4) model of Qwen/Qwen3.5-397B-A17B generated by intel/auto-round using model-free optimized RTN. Please follow the license of the original model.

NVFP4+ uses FP4 E2M1 elements for weights and activations, an FP8 E5M3 block scale with group size 16, and no tensor-wise global scale. Activations use dynamic per-block scaling.

Quantized (NVFP4+): all Linear layers in model.language_model.layers and mtp.layers, including routed experts, shared experts, attention projections, linear-attention projections, and the MTP FC/attention layers
Kept in BF16: router (mlp.gate), mlp.shared_expert_gate, in_proj_a / in_proj_b, lm_head, embeddings, and the vision tower
Kept in F32: linear_attn.norm
Not quantized: KV cache and attention softmax / SDPA

Property	Value
Base model	`Qwen/Qwen3.5-397B-A17B`
Quantization	NVFP4+ W4A4
Weight / activation element	FP4 E2M1
Block scale	FP8 E5M3
Group size	16
Tensor-wise global scale	None
Quantization algorithm	Model-free OPT-RTN
AutoRound version	0.16.0
Quantized modules	94,079
Quantized parameters	~400.75B
BF16/F32 parameters	~2.64B
Release size	~214.9 GiB

How to Use

Environment

The checkpoint uses the auto_round:llm_compressor_nvfp4_e5m3 packing format. Use an inference stack whose compressed-tensors / vLLM build supports NVFP4+ with E5M3 block scales; the checkpoint was produced with auto-round 0.16.0.

vLLM Usage

vllm serve intel-ai/Qwen3.5-397B-A17B-NVFP4-Plus-AutoRound

Evaluation

Recovery is the quantized score divided by the BF16 baseline.

Benchmark	BF16	NVFP4+	Recovery
GSM8K	97.65	97.12	99.46%
MMLU	88.56	87.90	99.25%
PIQA	83.03	82.48	99.34%
HellaSwag	74.23	72.67	97.90%
Average	85.87	85.04	99.04%

Average loss versus BF16: 0.96%.

Generate the Model

auto-round Qwen/Qwen3.5-397B-A17B \
    --model_free \
    --scheme NVFP4_E5M3 \
    --ignore_layers "visual,lm_head,embed_tokens,mlp.gate,mlp.shared_expert_gate,in_proj_a,in_proj_b" \
    --format auto_round \
    --output_dir ./Qwen3.5-397B-A17B-NVFP4-Plus

Ethical Considerations and Limitations

The model can produce factually incorrect output, and should not be relied on to produce factually accurate information. Because of the limitations of the pretrained model and the finetuning datasets, it is possible that this model could generate lewd, biased or otherwise offensive outputs. Therefore, before deploying any applications of the model, developers should perform safety testing.

Caveats and Recommendations

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Here are a couple of useful links to learn more about Intel's AI software:

Disclaimer

The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please consult an attorney before using this model for commercial purposes.

Cite

@article{cheng2023optimize,
  title={Optimize weight rounding via signed gradient descent for the quantization of llms},
  author={Cheng, Wenhua and Zhang, Weiwei and Shen, Haihao and Cai, Yiyang and He, Xin and Lv, Kaokao and Liu, Yi},
  journal={arXiv preprint arXiv:2309.05516},
  year={2023}
}

arxiv github

Signed-off-by: Kaokao, Lv kaokao.lv@intel.com

Downloads last month: 31

Safetensors

Model size

403B params

Tensor type

BF16

F32

Model tree for intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound

Base model

Qwen/Qwen3.5-397B-A17B

Quantized

(79)

this model

Paper for intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound

Optimize Weight Rounding via Signed Gradient Descent for the Quantization of LLMs

Paper • 2309.05516 • Published Sep 11, 2023 • 15