Instructions to use OpenGVLab/Mini-InternVL-Chat-2B-V1-5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenGVLab/Mini-InternVL-Chat-2B-V1-5 with Transformers:

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

pipe = pipeline("image-text-to-text", model="OpenGVLab/Mini-InternVL-Chat-2B-V1-5", trust_remote_code=True)
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 AutoModel
model = AutoModel.from_pretrained("OpenGVLab/Mini-InternVL-Chat-2B-V1-5", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use OpenGVLab/Mini-InternVL-Chat-2B-V1-5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenGVLab/Mini-InternVL-Chat-2B-V1-5"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenGVLab/Mini-InternVL-Chat-2B-V1-5",
		"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/OpenGVLab/Mini-InternVL-Chat-2B-V1-5

SGLang

How to use OpenGVLab/Mini-InternVL-Chat-2B-V1-5 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 "OpenGVLab/Mini-InternVL-Chat-2B-V1-5" \
    --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": "OpenGVLab/Mini-InternVL-Chat-2B-V1-5",
		"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 "OpenGVLab/Mini-InternVL-Chat-2B-V1-5" \
        --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": "OpenGVLab/Mini-InternVL-Chat-2B-V1-5",
		"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 OpenGVLab/Mini-InternVL-Chat-2B-V1-5 with Docker Model Runner:
```
docker model run hf.co/OpenGVLab/Mini-InternVL-Chat-2B-V1-5
```

Loading a 4-bit quantized version of Mini-InternVL-Chat-2B-V1-5 and run inference with transformers

by belofn - opened Oct 16, 2024

Discussion

belofn

Oct 16, 2024

•

edited Oct 16, 2024

we're loading and running the model with "Inference with Transformers" as per https://huggingface.co/OpenGVLab/Mini-InternVL-Chat-2B-V1-5#inference-with-transformers, all works well.

for the sake of performance, we quantized the model with AWQ, and would like to use the same approach as above. What should be the right way for loading the quantized model, while still using Inference with Transformers code?

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