Instructions to use OraRL/Video-ORA-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OraRL/Video-ORA-4B with Transformers:

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

pipe = pipeline("image-text-to-text", model="OraRL/Video-ORA-4B")
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, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("OraRL/Video-ORA-4B")
model = AutoModelForMultimodalLM.from_pretrained("OraRL/Video-ORA-4B", device_map="auto")
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 Settings

vLLM

How to use OraRL/Video-ORA-4B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OraRL/Video-ORA-4B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OraRL/Video-ORA-4B",
		"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/OraRL/Video-ORA-4B

SGLang

How to use OraRL/Video-ORA-4B 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 "OraRL/Video-ORA-4B" \
    --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": "OraRL/Video-ORA-4B",
		"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 "OraRL/Video-ORA-4B" \
        --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": "OraRL/Video-ORA-4B",
		"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 OraRL/Video-ORA-4B with Docker Model Runner:
```
docker model run hf.co/OraRL/Video-ORA-4B
```

Video-ORA-4B / code /tests /test_environment_spec.py

lyhisme

Add 4B model card, benchmark assets, and release code

0185029 verified about 11 hours ago

Raw

History Blame Contribute Delete

2.58 kB

	from __future__ import annotations

	from pathlib import Path

	import yaml

	RELEASE_ROOT = Path(__file__).resolve().parents[1]


	def test_h20_environment_uses_the_pinned_cuda_stack() -> None:
	environment = yaml.safe_load((RELEASE_ROOT / "environment.yml").read_text(encoding="utf-8"))
	dependencies = set(environment["dependencies"])

	assert environment["name"] == "orarl"
	assert "python=3.11" in dependencies
	assert "cuda-toolkit=12.9" in dependencies
	assert "ffmpeg=7" in dependencies
	assert "conda-forge" in environment["channels"]

	installer = (RELEASE_ROOT / "scripts" / "create_conda_env.sh").read_text(encoding="utf-8")
	assert "https://download.pytorch.org/whl/cu129" in installer
	assert "torch==2.10.0+cu129" in installer
	assert "vllm==0.19.1" in installer
	assert "flash-attn==2.8.3" in installer
	assert "install_conda_runtime_hook.sh" in installer
	assert "--evaluation-only" in installer
	# A single checkout installs the trainer; no external runtime root remains.
	assert "--runtime-root" not in installer
	assert "ORARL_RUNTIME_ROOT" not in installer

	validator = (RELEASE_ROOT / "scripts" / "check_environment.py").read_text(
	encoding="utf-8"
	)
	assert '"verl.trainer.main"' in validator
	assert "--evaluation-only" in validator

	runtime_hook = (
	RELEASE_ROOT / "scripts" / "install_conda_runtime_hook.sh"
	).read_text(encoding="utf-8")
	assert "activate.d/orarl-runtime.sh" in runtime_hook
	assert 'Path(sysconfig.get_path("purelib")) / "nvidia"' in runtime_hook


	def test_runtime_requirements_match_the_released_qwen35_stack() -> None:
	requirements = {
	line
	for line in (RELEASE_ROOT / "requirements-cu129.txt")
	.read_text(encoding="utf-8")
	.splitlines()
	if line and not line.startswith("#")
	}

	assert "transformers==5.5.4" in requirements
	assert "qwen-vl-utils[decord]==0.0.14" in requirements
	assert "torchcodec==0.10.0" in requirements
	assert "ray[default]==2.54.0" in requirements
	assert "flash-linear-attention==0.4.2" in requirements

	evaluator = (
	RELEASE_ROOT / "eval" / "task" / "temporal_grounding" / "eval_timelens_hf.py"
	).read_text(encoding="utf-8")
	assert 'setdefault("FORCE_QWENVL_VIDEO_READER", "decord")' in evaluator

	segmentation_evaluator = (
	RELEASE_ROOT / "eval" / "task" / "segmentation" / "eval_seg_vllm.py"
	).read_text(encoding="utf-8")
	assert 'setdefault("FORCE_QWENVL_VIDEO_READER", "decord")' in segmentation_evaluator