Instructions to use openbmb/MiniCPM-V-2_6-int4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use openbmb/MiniCPM-V-2_6-int4 with Transformers:

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

pipe = pipeline("image-text-to-text", model="openbmb/MiniCPM-V-2_6-int4", 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("openbmb/MiniCPM-V-2_6-int4", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use openbmb/MiniCPM-V-2_6-int4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "openbmb/MiniCPM-V-2_6-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": "openbmb/MiniCPM-V-2_6-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/openbmb/MiniCPM-V-2_6-int4

SGLang

How to use openbmb/MiniCPM-V-2_6-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 "openbmb/MiniCPM-V-2_6-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": "openbmb/MiniCPM-V-2_6-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 "openbmb/MiniCPM-V-2_6-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": "openbmb/MiniCPM-V-2_6-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 openbmb/MiniCPM-V-2_6-int4 with Docker Model Runner:
```
docker model run hf.co/openbmb/MiniCPM-V-2_6-int4
```

What method of quantization was used to make this model?

by CCRss - opened Aug 7, 2024

Discussion

CCRss

Aug 7, 2024

Great family models
I'm interested, is it just bnb quantization to 4bit? Or it's some close method that you can't publish.

tc-mb

OpenBMB org Aug 7, 2024

•

edited Aug 7, 2024

For the time being, it's just bnb.
Other quantization frameworks have difficulty supporting multimodal models, and we don't have enough energy to develop them yet.
Or you can use our version of gguf, which is also int4.
here: https://huggingface.co/openbmb/MiniCPM-V-2_6-gguf

CCRss

Aug 7, 2024

Thanks for the reply. 🙏 Sadly, maybe I will try to do AWQ quantization of this model if my supervisor accept it.

Aniel99

Aug 7, 2024

@CCRss I'm am also interested in AWQ/MLC quantization for running on Jetson devices. Maybe we can join effort to some point

CCRss

Aug 8, 2024

@Aniel99 I will be happy to. You can contact me via twitter. https://x.com/VladimirAlbrek1

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