Instructions to use Qwen/Qwen3-VL-8B-Instruct-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Qwen/Qwen3-VL-8B-Instruct-FP8 with Transformers:

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

pipe = pipeline("image-text-to-text", model="Qwen/Qwen3-VL-8B-Instruct-FP8")
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, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("Qwen/Qwen3-VL-8B-Instruct-FP8")
model = AutoModelForImageTextToText.from_pretrained("Qwen/Qwen3-VL-8B-Instruct-FP8")
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

vLLM

How to use Qwen/Qwen3-VL-8B-Instruct-FP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Qwen/Qwen3-VL-8B-Instruct-FP8"
# 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-VL-8B-Instruct-FP8",
		"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-VL-8B-Instruct-FP8

SGLang

How to use Qwen/Qwen3-VL-8B-Instruct-FP8 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-VL-8B-Instruct-FP8" \
    --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-VL-8B-Instruct-FP8",
		"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-VL-8B-Instruct-FP8" \
        --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-VL-8B-Instruct-FP8",
		"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-VL-8B-Instruct-FP8 with Docker Model Runner:
```
docker model run hf.co/Qwen/Qwen3-VL-8B-Instruct-FP8
```

missing <think> tags

by deepyes02 - opened Feb 9

Discussion

deepyes02

Feb 9

•

edited Feb 9

This specific qwen3-vl model seems to be responding in some form of jsond with a separate key 'reasoning' for streaming the reasoning data. However, the base model responds in tags inside the main "content" key, so I am confused because I want to use this vision model as a base model, and I loved the reasoning feature, but I am having difficulty implementing with libraries like ChatOpenAI or even ChatOllama with langchain.

Example request to

/v1/chat/completions:

curl --location 'http://xxx.xxx.x.x:11434/v1/chat/completions' \
--header 'Content-Type: application/json' \
--data '{
    "model": "qwen3:14b /qwen3-vl:8b",
    "messages": [
      {"role": "user", "content": "Explain quantum mechanics, and how it has helped us understand the world better."}
    ],
    "max_tokens": 100,
    "stream": true
}'

Response:

Qwen3 Text Responds with <think> tag inside the key content :

data: {"id":"chatcmpl-681","object":"chat.completion.chunk","created":1770621334,"model":"qwen3:14b","system_fingerprint":"fp_ollama","choices":[{"index":0,"delta":{"role":"assistant","content":"\u003cthink\u003e"},"finish_reason":null}]}

data: {"id":"chatcmpl-681","object":"chat.completion.chunk","created":1770621334,"model":"qwen3:14b","system_fingerprint":"fp_ollama","choices":[{"index":0,"delta":{"role":"assistant","content":"\n"},"finish_reason":null}]}

Qwen 3 Vision responds with thinking in a key reasoning :

data: {"id":"chatcmpl-298","object":"chat.completion.chunk","created":1770621544,"model":"qwen3-vl:8b","system_fingerprint":"fp_ollama","choices":[{"index":0,"delta":{"role":"assistant","content":"","reasoning":"Okay"},"finish_reason":null}]}

data: {"id":"chatcmpl-298","object":"chat.completion.chunk","created":1770621544,"model":"qwen3-vl:8b","system_fingerprint":"fp_ollama","choices":[{"index":0,"delta":{"role":"assistant","content":"","reasoning":","},"finish_reason":null}]}

I am having trouble working with two different types of response from the same version of model....

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