Instructions to use Qwen/Qwen3.5-27B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Qwen/Qwen3.5-27B with Transformers:

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

pipe = pipeline("image-text-to-text", model="Qwen/Qwen3.5-27B")
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("Qwen/Qwen3.5-27B")
model = AutoModelForMultimodalLM.from_pretrained("Qwen/Qwen3.5-27B", 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]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Qwen/Qwen3.5-27B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Qwen/Qwen3.5-27B"
# 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/Qwen3.5-27B",
		"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/Qwen3.5-27B

SGLang

How to use Qwen/Qwen3.5-27B 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/Qwen3.5-27B" \
    --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/Qwen3.5-27B",
		"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/Qwen3.5-27B" \
        --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/Qwen3.5-27B",
		"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/Qwen3.5-27B with Docker Model Runner:
```
docker model run hf.co/Qwen/Qwen3.5-27B
```

model overthinking

#20

by cse2011 - opened Feb 28

Discussion

cse2011

Feb 28

i even disabled thinking and still the model response is too slow compared to gpt-oss-120b
my cmd to start is

cmd = [
sys.executable, '-m', 'vllm.entrypoints.openai.api_server',
'--seed', str(self.cfg.seed),
'--model', self.cfg.model_path,
'--served-model-name', self.cfg.served_model_name,
'--tensor-parallel-size', '1',
'--max-num-seqs', str(self.cfg.batch_size),
'--gpu-memory-utilization', str(self.cfg.gpu_memory_utilization),
'--host', '0.0.0.0',
'--port', str(self.port),
'--dtype', self.cfg.dtype,
'--kv-cache-dtype', self.cfg.kv_cache_dtype,
'--max-model-len', str(self.cfg.context_tokens),
'--stream-interval', str(self.cfg.stream_interval),
# '--async-scheduling',
'--disable-log-stats',
'--enable-prefix-caching',
'--trust-remote-code',
'--enable-chunked-prefill',
'--language-model-only',
#'--speculative-config','{"method": "mtp", "num_speculative_tokens": 1}',
'--enforce-eager',
'--reasoning-parser', 'qwen3',
]
return subprocess.Popen(
cmd, stdout=self.log_file, stderr=subprocess.STDOUT, start_new_session=True
)

and to use
extra_body = {

        "chat_template_kwargs": {"enable_thinking": self.cfg.enable_thinking},
        "top_k": 20,
        
    }
    stream = self.client.chat.completions.create(
        model=self.model_name,
        messages=messages,
        max_tokens=max_tokens,
        temperature=self.cfg.temperature,
        top_p=0.95,
        presence_penalty=1.5,
        logprobs=False,
        stop=self.cfg.stop_strings,
        seed=seed,
        stream=True,
        extra_body=extra_body,
    )

innosynth

Mar 5

https://drive.google.com/file/d/1u53vaR-HOicoGsPcsIQx4IGbwhDSIr1h/view?usp=sharing

use vllm with latest build will resolve this issue on thinking control

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