Instructions to use Qwen/Qwen2-VL-72B-Instruct-GPTQ-Int4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Local Apps Settings

How to use Qwen/Qwen2-VL-72B-Instruct-GPTQ-Int4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Qwen/Qwen2-VL-72B-Instruct-GPTQ-Int4"
# 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-72B-Instruct-GPTQ-Int4",
		"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-72B-Instruct-GPTQ-Int4

SGLang

How to use Qwen/Qwen2-VL-72B-Instruct-GPTQ-Int4 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-72B-Instruct-GPTQ-Int4" \
    --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-72B-Instruct-GPTQ-Int4",
		"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-72B-Instruct-GPTQ-Int4" \
        --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-72B-Instruct-GPTQ-Int4",
		"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-72B-Instruct-GPTQ-Int4 with Docker Model Runner:
```
docker model run hf.co/Qwen/Qwen2-VL-72B-Instruct-GPTQ-Int4
```

Any one can use VLLM or any other engine support dynamic batch to run this with more than 1 GPU?

by bash99 - opened Sep 21, 2024

Discussion

bash99

Sep 21, 2024

I can run this with the example python code.

But vllm alway complained that "ValueError: The input size is not aligned with the quantized weight shape.", which seems is no solution from vllm side
https://github.com/vllm-project/vllm/issues/5675

Or can I run convert it to GPTQ use groupsize of 64, I'm not sure does vllm support it.

bash99

Sep 21, 2024

I've find https://qwen.readthedocs.io/en/latest/quantization/gptq.html , in the Troubleshooting section, it said You can pad the origin model then quantize it, which require a very large memory. Why don't just release one that works?

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