Instructions to use tencent/Penguin-VL-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tencent/Penguin-VL-2B with Transformers:

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

pipe = pipeline("text-generation", model="tencent/Penguin-VL-2B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("tencent/Penguin-VL-2B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use tencent/Penguin-VL-2B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tencent/Penguin-VL-2B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tencent/Penguin-VL-2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tencent/Penguin-VL-2B

SGLang

How to use tencent/Penguin-VL-2B 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 "tencent/Penguin-VL-2B" \
    --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": "tencent/Penguin-VL-2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "tencent/Penguin-VL-2B" \
        --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": "tencent/Penguin-VL-2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tencent/Penguin-VL-2B with Docker Model Runner:
```
docker model run hf.co/tencent/Penguin-VL-2B
```

Penguin-VL-2B / chat_template.json

lkeab

Transfer from pg-team/pg-vl-2b-hf

a10fac4 verified 3 months ago

raw

history blame contribute delete

1.97 kB

	{
	"chat_template": "\n{%- set identifier = 'im' %}\n{% for message in messages %}\n {% if message['role'] == 'stream' %}\n {% set identifier = 'stream' %}\n {% else %}\n {% set identifier = 'im' %}\n {% endif %}\n {% if message['role'] is not none %}\n {{- '<\|' + identifier + '_start\|>' + message['role'] + '\n' -}}\n {% endif %}\n {% if message['content'] is string %}\n {{- message['content'] + '<\|' + identifier + '_end\|>\n' -}}\n {% else %}\n {% for content in message['content'] %}\n {% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}\n {% if 'time' in content %}\n {{- 'Time ' + content['time'] \| round(1) \| string + 's: ' -}}\n {% endif %}\n {{- image_token + '\n' -}}\n {% elif content['type'] == 'video' or 'video' in content or 'video_url' in content %}\n {% for i in range(content['num_frames']) %}\n {% if 'timestamps' in content and content['timestamps']\|length > 0 %}\n {{- 'Time ' + content['timestamps'][i] \| round(1) \| string + 's:' -}}\n {% endif %}\n {% if i < content['num_frames'] - 1 %}\n {{- image_token + ',' -}}\n {% else %}\n {{- image_token + '\n' -}}\n {% endif %}\n {% endfor %}\n {% elif content['type'] == 'text' or 'text' in content %}\n {{- content['text'] -}}\n {% endif %}\n {% endfor %}\n {% if message['role'] is not none %}\n {{- '<\|' + identifier + '_end\|>\n' -}}\n {% endif %}\n {% endif %}\n{% endfor %}\n{% if add_generation_prompt %}\n {{- '<\|im_start\|>assistant\n' -}}\n {% if not add_think_prompt %}\n {{- '<think>\n\n</think>\n\n' -}}\n {% endif %}\n{% endif %}\n"
	}