Instructions to use fredaddy/Qwen-VL-7B-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fredaddy/Qwen-VL-7B-2 with Transformers:

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

pipe = pipeline("image-text-to-text", model="fredaddy/Qwen-VL-7B-2")
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("fredaddy/Qwen-VL-7B-2")
model = AutoModelForMultimodalLM.from_pretrained("fredaddy/Qwen-VL-7B-2", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use fredaddy/Qwen-VL-7B-2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fredaddy/Qwen-VL-7B-2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fredaddy/Qwen-VL-7B-2",
		"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/fredaddy/Qwen-VL-7B-2

SGLang

How to use fredaddy/Qwen-VL-7B-2 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 "fredaddy/Qwen-VL-7B-2" \
    --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": "fredaddy/Qwen-VL-7B-2",
		"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 "fredaddy/Qwen-VL-7B-2" \
        --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": "fredaddy/Qwen-VL-7B-2",
		"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 fredaddy/Qwen-VL-7B-2 with Docker Model Runner:
```
docker model run hf.co/fredaddy/Qwen-VL-7B-2
```

Qwen-VL-7B-2 / handler.py

fredaddy

Update handler.py

537f66a verified about 2 years ago

Raw History Blame Contribute Delete

2.7 kB

import torch
from PIL import Image
import base64
from io import BytesIO
from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
from qwen_vl_utils import process_vision_info

class EndpointHandler:
    def __init__(self, path=""):
        self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
        
        # Load model and processor
        self.model = Qwen2VLForConditionalGeneration.from_pretrained(
            path, torch_dtype=torch.float16 if self.device.type == "cuda" else torch.float32, device_map="auto"
        ).to(self.device)
        self.processor = AutoProcessor.from_pretrained(path)

    def __call__(self, data):
        # Extract image and text from the input data
        image_data = data.get("inputs", {}).get("image", "")
        text_prompt = data.get("inputs", {}).get("text", "")
    
        if not image_data or not text_prompt:
            return {"error": "Both 'image' and 'text' must be provided in the input data."}
    
        # Process the image data
        try:
            image_bytes = base64.b64decode(image_data)
            image = Image.open(BytesIO(image_bytes)).convert("RGB")
        except Exception as e:
            return {"error": f"Failed to process image data: {e}"}
    
        # Prepare the input in the format expected by the model
        messages = [
            {
                "role": "user",
                "content": [
                    {"type": "image", "image": image},
                    {"type": "text", "text": text_prompt},
                ],
            }
        ]
    
        # Process the input
        text = self.processor.apply_chat_template(
            messages, tokenize=False, add_generation_prompt=True
        )
        image_inputs, video_inputs = process_vision_info(messages)
        inputs = self.processor(
            text=[text],
            images=image_inputs,
            videos=video_inputs,
            padding=True,
            return_tensors="pt",
        )
    
        # Move inputs to the appropriate device
        inputs = {k: v.to(self.device) for k, v in inputs.items()}
    
        # Generate output
        with torch.no_grad():
            output_ids = self.model.generate(
                **inputs,
                max_new_tokens=2000,  # Increased from 128 to 2000
                num_return_sequences=1,
                do_sample=True,
                temperature=0.7,
                top_p=0.95
            )
    
        # Decode the output
        output_text = self.processor.batch_decode(
            output_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False
        )[0]
    
        return {"generated_text": output_text}

	import torch
	from PIL import Image
	import base64
	from io import BytesIO
	from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
	from qwen_vl_utils import process_vision_info

	class EndpointHandler:
	def __init__(self, path=""):
	self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

	# Load model and processor
	self.model = Qwen2VLForConditionalGeneration.from_pretrained(
	path, torch_dtype=torch.float16 if self.device.type == "cuda" else torch.float32, device_map="auto"
	).to(self.device)
	self.processor = AutoProcessor.from_pretrained(path)

	def __call__(self, data):
	# Extract image and text from the input data
	image_data = data.get("inputs", {}).get("image", "")
	text_prompt = data.get("inputs", {}).get("text", "")

	if not image_data or not text_prompt:
	return {"error": "Both 'image' and 'text' must be provided in the input data."}

	# Process the image data
	try:
	image_bytes = base64.b64decode(image_data)
	image = Image.open(BytesIO(image_bytes)).convert("RGB")
	except Exception as e:
	return {"error": f"Failed to process image data: {e}"}

	# Prepare the input in the format expected by the model
	messages = [
	{
	"role": "user",
	"content": [
	{"type": "image", "image": image},
	{"type": "text", "text": text_prompt},
	],
	}
	]

	# Process the input
	text = self.processor.apply_chat_template(
	messages, tokenize=False, add_generation_prompt=True
	)
	image_inputs, video_inputs = process_vision_info(messages)
	inputs = self.processor(
	text=[text],
	images=image_inputs,
	videos=video_inputs,
	padding=True,
	return_tensors="pt",
	)

	# Move inputs to the appropriate device
	inputs = {k: v.to(self.device) for k, v in inputs.items()}

	# Generate output
	with torch.no_grad():
	output_ids = self.model.generate(
	**inputs,
	max_new_tokens=2000, # Increased from 128 to 2000
	num_return_sequences=1,
	do_sample=True,
	temperature=0.7,
	top_p=0.95
	)

	# Decode the output
	output_text = self.processor.batch_decode(
	output_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False
	)[0]

	return {"generated_text": output_text}