Instructions to use SkyAsl/Nanbeige4.1-VLM-Base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SkyAsl/Nanbeige4.1-VLM-Base with Transformers:

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

pipe = pipeline("image-text-to-text", model="SkyAsl/Nanbeige4.1-VLM-Base", 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 NanbeigeVLM
model = NanbeigeVLM.from_pretrained("SkyAsl/Nanbeige4.1-VLM-Base", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use SkyAsl/Nanbeige4.1-VLM-Base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SkyAsl/Nanbeige4.1-VLM-Base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SkyAsl/Nanbeige4.1-VLM-Base",
		"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/SkyAsl/Nanbeige4.1-VLM-Base

SGLang

How to use SkyAsl/Nanbeige4.1-VLM-Base 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 "SkyAsl/Nanbeige4.1-VLM-Base" \
    --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": "SkyAsl/Nanbeige4.1-VLM-Base",
		"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 "SkyAsl/Nanbeige4.1-VLM-Base" \
        --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": "SkyAsl/Nanbeige4.1-VLM-Base",
		"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 SkyAsl/Nanbeige4.1-VLM-Base with Docker Model Runner:
```
docker model run hf.co/SkyAsl/Nanbeige4.1-VLM-Base
```

SkyAsl commited on Mar 15

Commit

a7afeba

verified ·

1 Parent(s): 47650fe

Upload modeling_nanbeige_vlm.py with huggingface_hub

Browse files

Files changed (1) hide show

modeling_nanbeige_vlm.py +3 -1

modeling_nanbeige_vlm.py CHANGED Viewed

@@ -82,7 +82,7 @@ class NanbeigeVLMModel(PreTrainedModel):
         # ── Vision tower (frozen at Stage 1) ──────────────────────────────
         self.vision_tower = SiglipVisionModel.from_pretrained(
-            config.vision_model_id, torch_dtype=torch.bfloat16
         )
         self.vision_tower.requires_grad_(False)
         vision_hidden_size = self.vision_tower.config.hidden_size
@@ -94,12 +94,14 @@ class NanbeigeVLMModel(PreTrainedModel):
                 trust_remote_code=True,
                 torch_dtype=torch.bfloat16,
                 attn_implementation="flash_attention_2",
             )
         except (ImportError, ValueError):
             self.language_model = AutoModelForCausalLM.from_pretrained(
                 config.llm_model_id,
                 trust_remote_code=True,
                 torch_dtype=torch.bfloat16,
             )
         self.language_model.requires_grad_(False)
         llm_hidden_size = self.language_model.config.hidden_size

         # ── Vision tower (frozen at Stage 1) ──────────────────────────────
         self.vision_tower = SiglipVisionModel.from_pretrained(
+            config.vision_model_id, torch_dtype=torch.bfloat16, device_map=None
         )
         self.vision_tower.requires_grad_(False)
         vision_hidden_size = self.vision_tower.config.hidden_size
                 trust_remote_code=True,
                 torch_dtype=torch.bfloat16,
                 attn_implementation="flash_attention_2",
+                device_map=None
             )
         except (ImportError, ValueError):
             self.language_model = AutoModelForCausalLM.from_pretrained(
                 config.llm_model_id,
                 trust_remote_code=True,
                 torch_dtype=torch.bfloat16,
+                device_map=None
             )
         self.language_model.requires_grad_(False)
         llm_hidden_size = self.language_model.config.hidden_size