Instructions to use Qwen/Qwen3.8-27B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Qwen/Qwen3.8-27B with Transformers:

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

pipe = pipeline("image-text-to-text", model="Qwen/Qwen3.8-27B")
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/Qwen3.8-27B")
model = AutoModelForMultimodalLM.from_pretrained("Qwen/Qwen3.8-27B", 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/Qwen3.8-27B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Qwen/Qwen3.8-27B"
# 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/Qwen3.8-27B",
		"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/Qwen3.8-27B

SGLang

How to use Qwen/Qwen3.8-27B 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/Qwen3.8-27B" \
    --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/Qwen3.8-27B",
		"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/Qwen3.8-27B" \
        --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/Qwen3.8-27B",
		"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/Qwen3.8-27B with Docker Model Runner:
```
docker model run hf.co/Qwen/Qwen3.8-27B
```

🏆 ROCM release v10.0 provided significant performance bonus

#186

by puchuu - opened Sep 4

Discussion

puchuu

Sep 4

•

edited Sep 4

Hello everyone.

I've just updated the Docker build script and tested the unsloth Qwen 3.8 27b Q8_0 with recent ROCM v10.0.
My dual 9060 XT 16GB setup received about +7-10 t/s output generation speed!
So I am recommending everyone to update to the recent ROCM v10.0 release.

Sharing with you the performance results:

0.18.662.485 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =   2560, progress = 0.13, t =   3.05 s / 838.13 tokens per second
0.19.348.985 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =   3072, progress = 0.16, t =   3.74 s / 822.39 tokens per second
...
0.46.940.087 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  19255, progress = 1.00, t =  31.29 s / 615.31 tokens per second
0.47.503.304 I slot print_timing: id  0 | task 0 | prompt processing, n_tokens =  19330, progress = 1.00, t =  32.03 s / 603.48 tokens per second
0.51.171.665 I slot print_timing: id  0 | task 0 | n_gen =    125, tg =  40.19 t/s, tg_3s =  40.52 t/s
0.54.188.662 I slot print_timing: id  0 | task 0 | n_gen =    283, tg =  46.21 t/s, tg_3s =  52.37 t/s
...
3.05.573.562 I slot print_timing: id  0 | task 0 | n_gen =   5775, tg =  42.00 t/s, tg_3s =  34.20 t/s
3.08.656.695 I slot print_timing: id  0 | task 0 | n_gen =   5902, tg =  41.98 t/s, tg_3s =  41.19 t/s

Previously on ROCM v7.14 I had about 35 t/s avg and up to 45 t/s output. Now I have 42 t/s avg and up to 55 t/s output.

The most important llama options are:

--spec-type "draft-mtp"
--spec-draft-n-max "4"
--split-mode "tensor"
--fit "off"
--n-gpu-layers "all"
--reasoning-effort "medium"

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