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

Libraries

How to use Qwen/Qwen3.6-35B-A3B with Transformers:

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

pipe = pipeline("image-text-to-text", model="Qwen/Qwen3.6-35B-A3B")
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("Qwen/Qwen3.6-35B-A3B")
model = AutoModelForMultimodalLM.from_pretrained("Qwen/Qwen3.6-35B-A3B", 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]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
AMD Developer Cloud
Local Apps Settings

vLLM

How to use Qwen/Qwen3.6-35B-A3B with vLLM:

Install from pip and serve model

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

SGLang

How to use Qwen/Qwen3.6-35B-A3B 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 "Qwen/Qwen3.6-35B-A3B" \
    --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": "Qwen/Qwen3.6-35B-A3B",
		"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 "Qwen/Qwen3.6-35B-A3B" \
        --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": "Qwen/Qwen3.6-35B-A3B",
		"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 Qwen/Qwen3.6-35B-A3B with Docker Model Runner:
```
docker model run hf.co/Qwen/Qwen3.6-35B-A3B
```

Running Qwen 3.6 35B-A3B on a single 8GB consumer laptop GPU (RTX 4060)

#58

by myeongjun77 - opened May 18

Discussion

myeongjun77

May 18

Hi Qwen team,

I packaged a local execution tool that runs Qwen 3.6 35B-A3B on an RTX 4060 Laptop (8GB VRAM, 32GB RAM) at warm decode 33-34 tok/s. Sharing the validation package in case it's useful or you'd like to verify it on your end.

Setup

Model: Qwen 3.6 35B-A3B GGUF Q4_K_M (obtained from upstream, weights NOT redistributed)
Hardware: RTX 4060 Laptop, 8GB VRAM, 32GB RAM, Windows 11
Backend: ik_llama.cpp fork (Qwen 3.6 arch patch)
Warm decode: 33-34 tok/s (conservative public claim: 20+ tok/s)
Smoke set: 10/10 multi-lingual pass

Validation package (single binary, no Python required):
https://github.com/JorrrrrdDin/RESEARCH_PAPERS/tree/main/Paper_16_Rotary_GPU_Local_MoE

Also verified on Gemma 4 26B-A4B (same A-3~4B MoE pattern) at 40+ tok/s on the same hardware.

The internal mechanism is not disclosed in this public release (patent filings pending; ORCID 0009-0006-9540-4666 for context). I'd appreciate any feedback from the Qwen team on what additional validation would be useful — happy to test on forthcoming larger MoE variants if you're interested.

Thanks for the model.

— Myeong Jun Jo

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