Instructions to use lmms-lab/LLaVA-OneVision-1.5-4B-Base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lmms-lab/LLaVA-OneVision-1.5-4B-Base with Transformers:

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

pipe = pipeline("image-text-to-text", model="lmms-lab/LLaVA-OneVision-1.5-4B-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 AutoModel
model = AutoModel.from_pretrained("lmms-lab/LLaVA-OneVision-1.5-4B-Base", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use lmms-lab/LLaVA-OneVision-1.5-4B-Base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lmms-lab/LLaVA-OneVision-1.5-4B-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": "lmms-lab/LLaVA-OneVision-1.5-4B-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/lmms-lab/LLaVA-OneVision-1.5-4B-Base

SGLang

How to use lmms-lab/LLaVA-OneVision-1.5-4B-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 "lmms-lab/LLaVA-OneVision-1.5-4B-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": "lmms-lab/LLaVA-OneVision-1.5-4B-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 "lmms-lab/LLaVA-OneVision-1.5-4B-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": "lmms-lab/LLaVA-OneVision-1.5-4B-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 lmms-lab/LLaVA-OneVision-1.5-4B-Base with Docker Model Runner:
```
docker model run hf.co/lmms-lab/LLaVA-OneVision-1.5-4B-Base
```

Improve model card: Add paper/code/demo links, sample usage, update title & citations

by nielsr HF Staff - opened Oct 5, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+398

-17

nielsr

Oct 5, 2025

This pull request significantly enhances the model card for LLaVA-OneVision-1.5 by:

Updating the main title to align with the official paper title: "LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training".
Adding prominent links to the Hugging Face paper page (https://huggingface.co/papers/2509.23661), the GitHub repository (https://github.com/EvolvingLMMs-Lab/LLaVA-OneVision-1.5), and the Hugging Face Space demo (https://huggingface.co/spaces/lmms-lab/LLaVA-OneVision-1.5) for easy access.
Including a "Quick Start with HuggingFace" section with a Python code snippet from the GitHub README to guide users on model inference.
Adding the language: en tag to the metadata for better discoverability.
Restructuring and expanding the model card content with detailed sections (Introduction, Models, Datasets, Evaluation Results, Training Guides, Roadmaps, Contributors, Citation, Acknowledgement) directly imported from the comprehensive GitHub README.
Updating the "Citation" section with a more complete BibTeX entry from the GitHub README, including related papers.

These updates aim to provide a more complete, accurate, and user-friendly documentation for the model on the Hugging Face Hub.

Improve model card: Add paper/code/demo links, sample usage, update title & citations31c1d620

xiangan

LMMs-Lab org Oct 5, 2025

Thank you so much for your help and support with the PR! I truly appreciate your effort and time in making this happen

xiangan changed pull request status to merged Oct 5, 2025

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