Instructions to use MBZUAI/ParallelTubeDecoding-Qwen3-VL-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MBZUAI/ParallelTubeDecoding-Qwen3-VL-4B with Transformers:

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

pipe = pipeline("image-text-to-text", model="MBZUAI/ParallelTubeDecoding-Qwen3-VL-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("MBZUAI/ParallelTubeDecoding-Qwen3-VL-4B")
model = AutoModelForMultimodalLM.from_pretrained("MBZUAI/ParallelTubeDecoding-Qwen3-VL-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 MBZUAI/ParallelTubeDecoding-Qwen3-VL-4B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MBZUAI/ParallelTubeDecoding-Qwen3-VL-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": "MBZUAI/ParallelTubeDecoding-Qwen3-VL-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/MBZUAI/ParallelTubeDecoding-Qwen3-VL-4B

SGLang

How to use MBZUAI/ParallelTubeDecoding-Qwen3-VL-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 "MBZUAI/ParallelTubeDecoding-Qwen3-VL-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": "MBZUAI/ParallelTubeDecoding-Qwen3-VL-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 "MBZUAI/ParallelTubeDecoding-Qwen3-VL-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": "MBZUAI/ParallelTubeDecoding-Qwen3-VL-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 MBZUAI/ParallelTubeDecoding-Qwen3-VL-4B with Docker Model Runner:
```
docker model run hf.co/MBZUAI/ParallelTubeDecoding-Qwen3-VL-4B
```

Update pipeline tag and add paper link

by nielsr HF Staff - opened about 10 hours ago

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

-6

Files changed (1) hide show

README.md +6 -6

README.md CHANGED Viewed

@@ -1,10 +1,10 @@
 ---
-license: apache-2.0
-library_name: transformers
-pipeline_tag: image-text-to-text
 base_model: Qwen/Qwen3-VL-4B-Instruct
 language:
 - en
 tags:
 - qwen3-vl
 - video
@@ -18,8 +18,8 @@ tags:
 # ParallelTubeDecoding-Qwen3-VL-4B
 This repository contains the merged Qwen3-VL-4B checkpoint for **Parallel
-Tube Decoding (PTD)** from *Locate Anything in Videos: Rethinking Efficient
-Generative Spatio-Temporal Video Grounding*.
 PTD first predicts the temporal interval of a queried event and then generates
 all time-conditioned spatial blocks in parallel. Decoupled Block Attention
@@ -179,4 +179,4 @@ occluded targets remain challenging.
 ## Acknowledgements
 This work builds on Qwen3-VL and the Qwen-VL-Series-Finetune training
-framework. See the code repository for full acknowledgements.

 ---
 base_model: Qwen/Qwen3-VL-4B-Instruct
 language:
 - en
+library_name: transformers
+license: apache-2.0
+pipeline_tag: video-text-to-text
 tags:
 - qwen3-vl
 - video
 # ParallelTubeDecoding-Qwen3-VL-4B
 This repository contains the merged Qwen3-VL-4B checkpoint for **Parallel
+Tube Decoding (PTD)** from [*Locate Anything in Videos: Rethinking Efficient
+Generative Spatio-Temporal Video Grounding*](https://huggingface.co/papers/2608.28192).
 PTD first predicts the temporal interval of a queried event and then generates
 all time-conditioned spatial blocks in parallel. Decoupled Block Attention
 ## Acknowledgements
 This work builds on Qwen3-VL and the Qwen-VL-Series-Finetune training
+framework. See the code repository for full acknowledgements.