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

Libraries

How to use fredaddy/MiniCPM-v-2_6 with Transformers:

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

pipe = pipeline("image-text-to-text", model="fredaddy/MiniCPM-v-2_6", 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("fredaddy/MiniCPM-v-2_6", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use fredaddy/MiniCPM-v-2_6 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fredaddy/MiniCPM-v-2_6"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fredaddy/MiniCPM-v-2_6",
		"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/fredaddy/MiniCPM-v-2_6

SGLang

How to use fredaddy/MiniCPM-v-2_6 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 "fredaddy/MiniCPM-v-2_6" \
    --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": "fredaddy/MiniCPM-v-2_6",
		"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 "fredaddy/MiniCPM-v-2_6" \
        --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": "fredaddy/MiniCPM-v-2_6",
		"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 fredaddy/MiniCPM-v-2_6 with Docker Model Runner:
```
docker model run hf.co/fredaddy/MiniCPM-v-2_6
```

fredaddy commited on Sep 17, 2024

Commit

830d07f

1 Parent(s): bb2a012

add custom handler

Browse files

Files changed (2) hide show

handler.py +1 -1
requirements.txt +0 -1

handler.py CHANGED Viewed

@@ -7,7 +7,7 @@ class EndpointHandler:
         self.model = AutoModel.from_pretrained(
             path,
             trust_remote_code=True,
-            attn_implementation='sdpa',
             torch_dtype=torch.float16
         )
         self.model = self.model.eval().cuda()

         self.model = AutoModel.from_pretrained(
             path,
             trust_remote_code=True,
+            attn_implementation='sdpa',  # Using sdpa instead of flash_attention_2
             torch_dtype=torch.float16
         )
         self.model = self.model.eval().cuda()

requirements.txt CHANGED Viewed

@@ -3,4 +3,3 @@ torch==2.1.2
 torchvision==0.16.2
 transformers==4.40.0
 sentencepiece==0.1.99
-flash-attn==2.3.6

 torchvision==0.16.2
 transformers==4.40.0
 sentencepiece==0.1.99