Instructions to use HuggingFaceTB/SmolVLM-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HuggingFaceTB/SmolVLM-Instruct with Transformers:

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

pipe = pipeline("image-text-to-text", model="HuggingFaceTB/SmolVLM-Instruct")
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 AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("HuggingFaceTB/SmolVLM-Instruct")
model = AutoModelForImageTextToText.from_pretrained("HuggingFaceTB/SmolVLM-Instruct")
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?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use HuggingFaceTB/SmolVLM-Instruct with vLLM:

Install from pip and serve model

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

SGLang

How to use HuggingFaceTB/SmolVLM-Instruct 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 "HuggingFaceTB/SmolVLM-Instruct" \
    --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": "HuggingFaceTB/SmolVLM-Instruct",
		"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 "HuggingFaceTB/SmolVLM-Instruct" \
        --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": "HuggingFaceTB/SmolVLM-Instruct",
		"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 HuggingFaceTB/SmolVLM-Instruct with Docker Model Runner:
```
docker model run hf.co/HuggingFaceTB/SmolVLM-Instruct
```

Xenova HF Staff commited on Apr 3, 2025

Commit

e6c9dea

verified ·

1 Parent(s): 4c2070c

Upload optimized language model w/ WebGPU-compatible GQA (#29)

Browse files

- Upload optimized language model w/ WebGPU-compatible GQA (9b303242a8f9d326545b843e1df7034817c2edd3)
- Update config.json (ea0c9de5c48509edf8e2b35059502ff66a17e18f)

Files changed (11) hide show

config.json +4 -0
onnx/decoder_model_merged.onnx +2 -2
onnx/decoder_model_merged.onnx_data +1 -1
onnx/decoder_model_merged_bnb4.onnx +2 -2
onnx/decoder_model_merged_fp16.onnx +2 -2
onnx/decoder_model_merged_fp16.onnx_data +2 -2
onnx/decoder_model_merged_int8.onnx +2 -2
onnx/decoder_model_merged_q4.onnx +2 -2
onnx/decoder_model_merged_q4f16.onnx +2 -2
onnx/decoder_model_merged_quantized.onnx +2 -2
onnx/decoder_model_merged_uint8.onnx +2 -2

@@ -179,6 +179,10 @@
       "embed_tokens": "auto",
       "vision_encoder": "auto",
       "decoder_model_merged": "q4"
     }
   },
   "use_cache": true,

       "embed_tokens": "auto",
       "vision_encoder": "auto",
       "decoder_model_merged": "q4"
+    },
+    "use_external_data_format": {
+      "decoder_model_merged.onnx": true,
+      "decoder_model_merged_fp16.onnx": true
     }
   },
   "use_cache": true,

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a27ef6fe177d3109e0913c63da7a4b0f2791fab95da3e5f91b31ba6e03115385
-size 126930

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d530c318000311b2697d0b891ef46c69f9e9c89688761e043654d08a3cca376c
 size 6849724416

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:39974ffc8a05f4de601005dc555d326f3dd2744ffd544e2892ad065fe25b2b8a
-size 967330291

 version https://git-lfs.github.com/spec/v1
+oid sha256:a6a5013f7392cb1ab70de86049bdeaf4522599e682e51eeaf1ad22d0ebfffa44
+size 167944

 version https://git-lfs.github.com/spec/v1
+oid sha256:e347876c58484dc6c51422ce04265e6054624754f1f8bb9cc7ac6076e55856bf
+size 967377227

 version https://git-lfs.github.com/spec/v1
+oid sha256:ec9bfa341a42d5060d69573f5dab1c8be24d9d8cc78af8649c6b077223dcb0c3
+size 169023

 version https://git-lfs.github.com/spec/v1
+oid sha256:2cc57156f23eeb36f3a8ed08504b2d050f8d221c1a67dcfc1eea089f93154437
+size 3424862208

 version https://git-lfs.github.com/spec/v1
+oid sha256:861dcec66dc59cd9e7f2622cf28b60917020f23ac43f11ab179893773da5dbda
+size 1716222232

 version https://git-lfs.github.com/spec/v1
+oid sha256:79e171c9285756185c74ed67e0f598d97bae71d2996f3752fb62dd08db4cbba6
+size 1074331012

 version https://git-lfs.github.com/spec/v1
+oid sha256:247de27f420790b553b94b29713c56278aedca18b8d68d63dd2e7d72cca397ef
+size 965077981

 version https://git-lfs.github.com/spec/v1
+oid sha256:9614d219c614af85de470c8870be65e8a3a1d00a8163fb3df8abb6cecb612399
+size 1716222303