Instructions to use InternRobotics/VLAC-Cut with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use InternRobotics/VLAC-Cut with Transformers:

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

pipe = pipeline("image-text-to-text", model="InternRobotics/VLAC-Cut")
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("InternRobotics/VLAC-Cut")
model = AutoModelForMultimodalLM.from_pretrained("InternRobotics/VLAC-Cut", 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 InternRobotics/VLAC-Cut with vLLM:

Install from pip and serve model

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

SGLang

How to use InternRobotics/VLAC-Cut 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 "InternRobotics/VLAC-Cut" \
    --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": "InternRobotics/VLAC-Cut",
		"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 "InternRobotics/VLAC-Cut" \
        --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": "InternRobotics/VLAC-Cut",
		"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 InternRobotics/VLAC-Cut with Docker Model Runner:
```
docker model run hf.co/InternRobotics/VLAC-Cut
```

futurefantasy commited on Jun 23

Commit

adc84f9

verified ·

1 Parent(s): 4008484

Refine README inference usage and argument descriptions

Browse files

Files changed (2) hide show

README.md +12 -1
quick_start/README.md +15 -3

README.md CHANGED Viewed

@@ -49,10 +49,21 @@ Run progress inference on an arbitrary video:
 ```bash
 python quick_start/run_example.py \
   --video-path /path/to/video.mp4 \
-  --task-instruction "把洋葱放进快递箱里。"
 ```
 Render a prediction-only preview video from the JSONL output:
 ```bash

 ```bash
 python quick_start/run_example.py \
+  --model-path /path/to/VLAC-Cut-Qwen3VL-30B-A3B-Progress \
   --video-path /path/to/video.mp4 \
+  --task-instruction "<natural-language task instruction>" \
+  --task-plan $'<optional step-by-step task plan>' \
+  --output-jsonl quick_start/outputs/video.jsonl
 ```
+- `--video-path`: required input video.
+- `--task-instruction`: task description, unless `--prompt` is used instead.
+- `--task-plan`: optional step-by-step plan text for the default `chunk_all` prompt.
+- `--model-path`: optional. Defaults to the parent directory of `quick_start/`.
+- `--output-jsonl`: optional. Defaults to `quick_start/outputs/<video_stem>.jsonl`.
+- `--prompt`: optional full prompt override.
+- `--sample-hz`, `--max-new-tokens`: optional inference controls.
 Render a prediction-only preview video from the JSONL output:
 ```bash

quick_start/README.md CHANGED Viewed

@@ -23,10 +23,24 @@ Run inference:
 ```bash
 python quick_start/run_example.py \
   --video-path /path/to/video.mp4 \
-  --task-instruction "把洋葱放进快递箱里。"
 ```
 Render a preview video:
 ```bash
@@ -35,8 +49,6 @@ python quick_start/render_prediction_video.py \
   --output-video quick_start/outputs/video_pred_progress.mp4
 ```
-If the model directory is not the parent of `quick_start/`, pass `--model-path`.
 ## Bundled Examples
 `example_01`

 ```bash
 python quick_start/run_example.py \
+  --model-path /path/to/VLAC-Cut-Qwen3VL-30B-A3B-Progress \
   --video-path /path/to/video.mp4 \
+  --task-instruction "<natural-language task instruction>" \
+  --task-plan $'<optional step-by-step task plan>' \
+  --output-jsonl quick_start/outputs/video.jsonl
 ```
+Important arguments:
+- `--video-path`: required. Input video path.
+- `--task-instruction`: optional if `--prompt` is provided. Natural-language task description.
+- `--task-plan`: optional. Multi-line plan text appended after `--task-instruction` when building the default `chunk_all` prompt.
+- `--model-path`: optional. Defaults to the parent directory of `quick_start/`.
+- `--output-jsonl`: optional. Defaults to `quick_start/outputs/<video_stem>.jsonl`.
+- `--sample-hz`: optional. Video sampling rate for inference. Default: `2.0`.
+- `--max-new-tokens`: optional. Generation cap. Default: `1024`.
+- `--prompt`: optional full prompt override. If set, `--task-instruction` and `--task-plan` are ignored.
 Render a preview video:
 ```bash
   --output-video quick_start/outputs/video_pred_progress.mp4
 ```
 ## Bundled Examples
 `example_01`