Instructions to use OpenGVLab/Mini-InternVL2-4B-DA-Medical with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenGVLab/Mini-InternVL2-4B-DA-Medical with Transformers:

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

pipe = pipeline("image-text-to-text", model="OpenGVLab/Mini-InternVL2-4B-DA-Medical", 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-InternVL2-4B-DA-Medical", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use OpenGVLab/Mini-InternVL2-4B-DA-Medical with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenGVLab/Mini-InternVL2-4B-DA-Medical"
# 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-InternVL2-4B-DA-Medical",
		"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-InternVL2-4B-DA-Medical

SGLang

How to use OpenGVLab/Mini-InternVL2-4B-DA-Medical 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-InternVL2-4B-DA-Medical" \
    --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-InternVL2-4B-DA-Medical",
		"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-InternVL2-4B-DA-Medical" \
        --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-InternVL2-4B-DA-Medical",
		"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-InternVL2-4B-DA-Medical with Docker Model Runner:
```
docker model run hf.co/OpenGVLab/Mini-InternVL2-4B-DA-Medical
```

Wrong tokenizer given "expected str, bytes or os.PathLike object, not NoneType"

by DasChameleon - opened Feb 2, 2025

Discussion

DasChameleon

Feb 2, 2025

Running both the 4B and 2B versions raises Exceptions for me "expected str, bytes or os.PathLike object, not NoneType" when trying to load the tokenizer.
I think the issue is that the LlamaTokenizer expects a tokenizer.model, but this repo contains a tokenizer.json instead.

Here's how I solved it:
- Run the model once, so it has all the files downloaded
- Copy over tokenizer.model (in snapshots)and the file it points to (in blobs) (the file in blobs first, so the pointer works) from the non-adapted version of InternVL2 (either the 2B or 4B version)
- Delete tokenizer.model in the .no_exist directory (it was cached as missing during the initial download run)

Results seem to be good, but if you could add the "real" tokenizer.model, that would be great. Thanks!

yuchengc

Mar 1, 2025

I cannot find the tokenizer.model in snapshots file.

DasChameleon

Mar 1, 2025

They are the 2B or 4B paths I linked

yuchengc

Mar 1, 2025

I use 4B but couldn't find the tokenizer.model in snapshots file, In your solution, the second item "Copy over tokenizer.model (in snapshots)and the file it points to (in blobs) (the file in blobs first, so the pointer works) from the non-adapted version of InternVL2 (either the 2B or 4B version)", I have no idea what to copy

DasChameleon

Mar 18, 2025

https://huggingface.co/OpenGVLab/InternVL2-4B/blob/main/tokenizer.model

For example this

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