Instructions to use unsloth/Qwen3.6-35B-A3B-NVFP4-Fast with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use unsloth/Qwen3.6-35B-A3B-NVFP4-Fast with Transformers:

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

pipe = pipeline("image-text-to-text", model="unsloth/Qwen3.6-35B-A3B-NVFP4-Fast")
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)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("unsloth/Qwen3.6-35B-A3B-NVFP4-Fast")
model = AutoModelForMultimodalLM.from_pretrained("unsloth/Qwen3.6-35B-A3B-NVFP4-Fast", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use unsloth/Qwen3.6-35B-A3B-NVFP4-Fast with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "unsloth/Qwen3.6-35B-A3B-NVFP4-Fast"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "unsloth/Qwen3.6-35B-A3B-NVFP4-Fast",
		"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/unsloth/Qwen3.6-35B-A3B-NVFP4-Fast

SGLang

How to use unsloth/Qwen3.6-35B-A3B-NVFP4-Fast 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 "unsloth/Qwen3.6-35B-A3B-NVFP4-Fast" \
    --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": "unsloth/Qwen3.6-35B-A3B-NVFP4-Fast",
		"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 "unsloth/Qwen3.6-35B-A3B-NVFP4-Fast" \
        --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": "unsloth/Qwen3.6-35B-A3B-NVFP4-Fast",
		"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"
						}
					}
				]
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use unsloth/Qwen3.6-35B-A3B-NVFP4-Fast with Docker Model Runner:
```
docker model run hf.co/unsloth/Qwen3.6-35B-A3B-NVFP4-Fast
```

Can't load the model in 32G VRAM (vLLM)

by Duonglv - opened Jul 15

Discussion

Duonglv

Jul 15

•

edited Jul 15

Hello,
I have tried to load the model with 2 RTX 5060 Ti in the vLLM. I have 32G system Ram.
I saw that it loaded ~11G vRam each, full 32G system Ram then the PC hanged.
I tried to increase the swap size to 32G. It loaded for a while then the system killed it.
Did I miss something? what is the correct way to load this model?

My system loads well with this model: nvidia/Qwen3.6-35B-A3B-NVFP4

Duonglv

Jul 17

•

edited Jul 19

It seems the bottleneck is in the loading phase, not the inference phase.
We need a big amout of system Ram to load the model before compressing it to GPU Ram.
So, these nvfp4 models are more suitable for DGX Spark pc.

tutugreen

Aug 1

I was able to load the model with the full 262K context length on a single RTX 5090 32G VRAM without any issues (Fast or Non-Fast). I’ve posted my configuration and test results here:

https://huggingface.co/unsloth/Qwen3.6-35B-A3B-NVFP4/discussions/8

Duonglv

Aug 2

The reason in my case was pretty simple. I needed to run it inside Docker, not directly like vllm serve ...
Thanks, @tutugreen ! Your comment made me try a different approach, and that solved the problem.

Duonglv changed discussion status to closed Aug 2

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment