Instructions to use Qwen/Qwen2-VL-7B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Qwen/Qwen2-VL-7B-Instruct with Transformers:

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

pipe = pipeline("image-text-to-text", model="Qwen/Qwen2-VL-7B-Instruct")
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/Qwen2-VL-7B-Instruct")
model = AutoModelForMultimodalLM.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", 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 Qwen/Qwen2-VL-7B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Qwen/Qwen2-VL-7B-Instruct"
# 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/Qwen2-VL-7B-Instruct",
		"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/Qwen2-VL-7B-Instruct

SGLang

How to use Qwen/Qwen2-VL-7B-Instruct 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/Qwen2-VL-7B-Instruct" \
    --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/Qwen2-VL-7B-Instruct",
		"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/Qwen2-VL-7B-Instruct" \
        --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/Qwen2-VL-7B-Instruct",
		"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/Qwen2-VL-7B-Instruct with Docker Model Runner:
```
docker model run hf.co/Qwen/Qwen2-VL-7B-Instruct
```

You are attempting to use Flash Attention 2.0 without specifying a torch dtype.

#23

by chrislevy - opened Sep 7, 2024

Discussion

chrislevy

Sep 7, 2024

I am specifying the torch_dtype.

from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
import torch

# We recommend enabling flash_attention_2 for better acceleration and memory saving, especially in multi-image and video scenarios.

model = Qwen2VLForConditionalGeneration.from_pretrained(
    "Qwen/Qwen2-VL-7B-Instruct",
    torch_dtype=torch.bfloat16,
    attn_implementation="flash_attention_2",
    device_map="auto",
)
# default processor
processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")

But it says during inference

You are attempting to use Flash Attention 2.0 without specifying a torch dtype. This might lead to unexpected behaviour

Inference works. Just not really sure under the hood if everything is running properly.

I have installed flash attention in my container

cuda_version = "12.4.0"  # should be no greater than host CUDA version
flavor = "devel"  #  includes full CUDA toolkit
operating_sys = "ubuntu22.04"
tag = f"{cuda_version}-{flavor}-{operating_sys}"
image = (
    modal.Image.from_registry(f"nvidia/cuda:{tag}", add_python="3.11")
    .apt_install("git")
    .pip_install(
        "ninja",  # required to build flash-attn
        "packaging",  # required to build flash-attn
        "wheel",  # required to build flash-attn
    )
    .run_commands(
        "pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu124",
        "pip install git+https://github.com/huggingface/transformers",
        "pip install accelerate",
        "pip install qwen-vl-utils",
        "pip install python-dotenv",
        f'huggingface-cli login --token {os.environ["HUGGING_FACE_ACCESS_TOKEN"]}',
    )
    .run_commands("pip install flash-attn --no-build-isolation")
)

chrislevy

Sep 7, 2024

Maybe just a transformers thing ? https://github.com/huggingface/transformers/issues/28052

nbroad

Sep 7, 2024

It's probably because you didn't set torch_dtype in the vision config.

See this line here

If you run model.config.vision_config, what is the value for torch_dtype?

You can probably fix this by doing

model = Qwen2VLForConditionalGeneration.from_pretrained(
    "Qwen/Qwen2-VL-7B-Instruct",
    torch_dtype=torch.bfloat16,
    attn_implementation="flash_attention_2",
    device_map="auto",
    vision_config={"torch_dtype": torch.bfloat16}
)

chrislevy

Sep 7, 2024

Thanks for the suggestion! It is still logging that to the screen. Not sure if this warning can be ignored or if it really means it's not utilizing the flash attention.

nbroad

Sep 7, 2024

•

edited Sep 7, 2024

The warning can be ignored. Flash attention 2 can only work on bfloat16 or float16, and it will result in an error if you are not using those dtypes. All of the weights are in bfloat16 in the repo, so they should stay that way when you call from_pretrained. Still, it's annoying that it comes up. There is some work being done to improve the attn_implementation for composite models like Qwen2VL, so hopefully this will be addressed.

If you want to be absolutely sure, you can do attn_implementation="eager" and check if the memory consumption is higher and the speed is slower

chrislevy

Sep 7, 2024

Okay thanks for the quick replies!

chrislevy changed discussion status to closed Sep 7, 2024

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