Instructions to use ByteDance/Sa2VA-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ByteDance/Sa2VA-1B with Transformers:

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

pipe = pipeline("image-text-to-text", model="ByteDance/Sa2VA-1B", 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("ByteDance/Sa2VA-1B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ByteDance/Sa2VA-1B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ByteDance/Sa2VA-1B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ByteDance/Sa2VA-1B",
		"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/ByteDance/Sa2VA-1B

SGLang

How to use ByteDance/Sa2VA-1B 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 "ByteDance/Sa2VA-1B" \
    --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": "ByteDance/Sa2VA-1B",
		"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 "ByteDance/Sa2VA-1B" \
        --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": "ByteDance/Sa2VA-1B",
		"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 ByteDance/Sa2VA-1B with Docker Model Runner:
```
docker model run hf.co/ByteDance/Sa2VA-1B
```

Demo

by merve - opened Jan 9, 2025

Discussion

merve

Jan 9, 2025

Hello @Xiangtai congratulations on the release!

Would you like to build a demo to any of the models (even 8B) and we can provide zero A100 grant for it?

Also the model tree seems a bit off, I think base models should be only the following: InternVL2.0-1B Qwen2-0.5B-Instruct they're already mapped to their own base models so you don't have to add them. I opened a PR here https://huggingface.co/ByteDance/Sa2VA-1B/discussions/1 I will open the same for other ones.

merve

Jan 9, 2025

Also read the paper, it seems you have annotations, are you planning to release them as well?

wjbmattingly

Jan 12, 2025

•

edited Jan 12, 2025

I'd like to see both of these as well. @merve thanks for sharing it on LinkedIn! I did a few tests and it works remarkably well on HTR. It even does well with some medieval scripts, though not as well as Qwen 2 VL.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment