Instructions to use openbmb/MiniCPM-V-4_5-AWQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use openbmb/MiniCPM-V-4_5-AWQ with Transformers:

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

pipe = pipeline("image-text-to-text", model="openbmb/MiniCPM-V-4_5-AWQ", 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-4_5-AWQ", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use openbmb/MiniCPM-V-4_5-AWQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "openbmb/MiniCPM-V-4_5-AWQ"
# 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-4_5-AWQ",
		"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-4_5-AWQ

SGLang

How to use openbmb/MiniCPM-V-4_5-AWQ 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-4_5-AWQ" \
    --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-4_5-AWQ",
		"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-4_5-AWQ" \
        --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-4_5-AWQ",
		"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-4_5-AWQ with Docker Model Runner:
```
docker model run hf.co/openbmb/MiniCPM-V-4_5-AWQ
```

MiniCPM-V-4_5-AWQ / tokenization_minicpmv_fast.py

tc-mb

Initial commit: MiniCPM-V-4_5 model

ec15ed2 verified 9 months ago

raw

history blame contribute delete

1.65 kB

	from transformers import Qwen2TokenizerFast


	class MiniCPMVTokenizerFast(Qwen2TokenizerFast):
	def __init__(self, **kwargs):
	super().__init__(**kwargs)
	self.im_start = "<image>"
	self.im_end = "</image>"
	self.ref_start = "<ref>"
	self.ref_end = "</ref>"
	self.box_start = "<box>"
	self.box_end = "</box>"
	self.quad_start = "<quad>"
	self.quad_end = "</quad>"
	self.slice_start = "<slice>"
	self.slice_end = "</slice>"
	self.im_id_start = "<image_id>"
	self.im_id_end = "</image_id>"

	@property
	def eos_id(self):
	return self.eos_token_id

	@property
	def bos_id(self):
	return self.bos_token_id

	@property
	def unk_id(self):
	return self.unk_token_id

	@property
	def im_start_id(self):
	return self.convert_tokens_to_ids(self.im_start)

	@property
	def im_end_id(self):
	return self.convert_tokens_to_ids(self.im_end)

	@property
	def slice_start_id(self):
	return self.convert_tokens_to_ids(self.slice_start)

	@property
	def slice_end_id(self):
	return self.convert_tokens_to_ids(self.slice_end)

	@property
	def im_id_start_id(self):
	return self.convert_tokens_to_ids(self.im_id_start)

	@property
	def im_id_end_id(self):
	return self.convert_tokens_to_ids(self.im_id_end)

	@property
	def newline_id(self):
	return self.convert_tokens_to_ids('\n')

	@staticmethod
	def escape(text: str) -> str:
	return text

	@staticmethod
	def unescape(text: str) -> str:
	return text