Instructions to use ob11/Qwen-VL-PRM-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ob11/Qwen-VL-PRM-3B with Transformers:

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

pipe = pipeline("image-text-to-text", model="ob11/Qwen-VL-PRM-3B")
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("ob11/Qwen-VL-PRM-3B")
model = AutoModelForMultimodalLM.from_pretrained("ob11/Qwen-VL-PRM-3B")
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 ob11/Qwen-VL-PRM-3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ob11/Qwen-VL-PRM-3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ob11/Qwen-VL-PRM-3B",
		"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/ob11/Qwen-VL-PRM-3B

SGLang

How to use ob11/Qwen-VL-PRM-3B 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 "ob11/Qwen-VL-PRM-3B" \
    --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": "ob11/Qwen-VL-PRM-3B",
		"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 "ob11/Qwen-VL-PRM-3B" \
        --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": "ob11/Qwen-VL-PRM-3B",
		"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 ob11/Qwen-VL-PRM-3B with Docker Model Runner:
```
docker model run hf.co/ob11/Qwen-VL-PRM-3B
```

nielsr HF Staff commited on Oct 3, 2025

Commit

400dfa8

verified ·

1 Parent(s): 99d3938

Improve model card: Add pipeline tag, correct license, enhance summary

Browse files

This PR improves the model card for `ob11/Qwen-VL-PRM-3B` by:

* Adding the `pipeline_tag: image-text-to-text` to better categorize the model on the Hub.
* Correcting the spelling of the `licence` metadata key to `license: apache-2.0`.
* Enhancing the "Model Summary" section with more descriptive information derived from the paper's abstract.
* Removing a redundant `

Files changed (1) hide show

README.md +20 -19

README.md CHANGED Viewed

@@ -1,15 +1,16 @@
 ---
 base_model: Qwen/Qwen2.5-VL-3B-Instruct
-library_name: transformers
-model_name: ob11/Qwen-VL-PRM-3B
-licence: apache-2.0
 datasets:
 - ob11/VL-PRM300K-V1-train
 ---
 # Model Summary
-> Qwen-VL-PRM-3B is a process reward model finetuned from Qwen2.5-3B-Instruct on approximately 300,000 examples. It demonstrates strong test-time scaling performance improvements on various advanced multimodal reasoning benchmarks when used with Qwen2.5-VL and Gemma-3 models despite being trained mainly on abstract reasoning datasets and elementary reasoning datasets.
 - **Logs:** https://wandb.ai/aisg-arf/multimodal-reasoning/runs/pnsncs80
 - **Repository:** https://github.com/theogbrand/vlprm
@@ -28,23 +29,23 @@ The model usage is documented [here](https://github.com/theogbrand/vlprm/blob/ma
 | o3 | 82.9 | 84.1 | 62.3 | 86.8 | -- | -- |
 ### Qwen-2.5-VL Family
 | Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |
-|-------|------|-----------|---------------|-----------|------------|---------|
-| **Qwen-2.5-VL-3B** | 51.7 | 34.5 | 25.7 | 60.0 | 21.2 | 38.6 |
-| + VL-PRM-7B | 53.7 (+2.0) | 44.9 (+10.5) | 28.3 (+2.6) | 64.1 (+4.1) | 21.8 (+0.6) | 42.6 (+4.0) |
-| **Qwen-2.5-VL-7B** | 55.0 | 48.0 | 29.1 | 67.8 | 24.2 | 44.8 |
-| + VL-PRM-3B | 57.6 (+2.6) | 55.5 (+7.5) | 33.8 (+4.7) | 70.0 (+2.2) | 26.1 (+1.9) | 48.6 (+3.6) |
-| + VL-PRM-7B | 57.4 (+2.4) | 54.8 (+6.8) | 35.3 (+6.2) | 71.0 (+3.2) | 26.2 (+2.0) | 48.9 (+4.1) |
-| **Qwen-2.5-VL-32B** | 66.0 | 46.2 | 26.9 | 76.9 | 36.7 | 50.5 |
-| + VL-PRM-3B | 67.0 (+1.0) | 67.1 (+20.8) | 41.6 (+14.7) | 77.7 (+0.8) | 40.5 (+3.8) | 58.7 (+8.2) |
-| + VL-PRM-7B | 67.6 (+1.6) | 66.8 (+20.6) | 44.2 (+17.3) | 78.3 (+1.4) | 40.1 (+3.2) | 59.4 (+8.9) |
 ### Gemma-3 Family
 | Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |
-|-------|------|-----------|---------------|-----------|------------|---------|
-| **Gemma-3-12B** | 57.6 | 45.0 | 29.1 | 58.9 | 28.1 | 43.7 |
-| + VL-PRM-3B | 60.4 (+2.8) | 57.7 (+12.7) | 39.7 (+10.6) | 60.3 (+1.4) | 33.8 (+5.7) | 50.4 (+6.7) |
-| + VL-PRM-7B | 60.2 (+2.6) | 59.0 (+12.0) | 41.1 (+4.5) | 63.3 (+4.4) | 33.9 (+5.8) | 51.5 (+7.8) |
-| **Gemma-3-27B** | 62.9 | 50.8 | 29.9 | 61.6 | 32.4 | 47.5 |
-| + VL-PRM-3B | 65.5 (+2.6) | 67.4 (+16.6) | 40.3 (+10.4) | 65.4 (+3.8) | 39.8 (+7.4) | 55.7 (+8.2) |
 | + VL-PRM-7B | 64.5 (+1.6) | 67.6 (+16.8) | 41.1 (+11.2) | 65.2 (+3.6) | 40.9 (+8.5) | 55.9 (+8.4) |
 ### Framework versions

 ---
 base_model: Qwen/Qwen2.5-VL-3B-Instruct
 datasets:
 - ob11/VL-PRM300K-V1-train
+library_name: transformers
+model_name: ob11/Qwen-VL-PRM-3B
+license: apache-2.0
+pipeline_tag: image-text-to-text
 ---
 # Model Summary
+Qwen-VL-PRM-3B is a process reward model fine-tuned from Qwen2.5-3B-Instruct on approximately 300,000 examples. It provides step-level supervision to improve the reliability of reasoning in large language models. The model introduces a hybrid data synthesis framework that combines MCTS with judgments from a strong VLM, and proposes perception-focused supervision. It systematically evaluates diverse strategies for dataset construction, training, and test-time scaling. This model demonstrates strong test-time scaling performance improvements on various advanced multimodal reasoning benchmarks including MMMU, PuzzleVQA, AlgoPuzzleVQA, MathVista, and MathVision, when used with Qwen2.5-VL and Gemma-3 models, despite being trained mainly on abstract reasoning datasets and elementary reasoning datasets.
 - **Logs:** https://wandb.ai/aisg-arf/multimodal-reasoning/runs/pnsncs80
 - **Repository:** https://github.com/theogbrand/vlprm
 | o3 | 82.9 | 84.1 | 62.3 | 86.8 | -- | -- |
 ### Qwen-2.5-VL Family
 | Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |
+|-------|------|-----------|---------------|-----------|------------|---------|\
+| **Qwen-2.5-VL-3B** | 51.7 | 34.5 | 25.7 | 60.0 | 21.2 | 38.6 |\
+| + VL-PRM-7B | 53.7 (+2.0) | 44.9 (+10.5) | 28.3 (+2.6) | 64.1 (+4.1) | 21.8 (+0.6) | 42.6 (+4.0) |\
+| **Qwen-2.5-VL-7B** | 55.0 | 48.0 | 29.1 | 67.8 | 24.2 | 44.8 |\
+| + VL-PRM-3B | 57.6 (+2.6) | 55.5 (+7.5) | 33.8 (+4.7) | 70.0 (+2.2) | 26.1 (+1.9) | 48.6 (+3.6) |\
+| + VL-PRM-7B | 57.4 (+2.4) | 54.8 (+6.8) | 35.3 (+6.2) | 71.0 (+3.2) | 26.2 (+2.0) | 48.9 (+4.1) |\
+| **Qwen-2.5-VL-32B** | 66.0 | 46.2 | 26.9 | 76.9 | 36.7 | 50.5 |\
+| + VL-PRM-3B | 67.0 (+1.0) | 67.1 (+20.8) | 41.6 (+14.7) | 77.7 (+0.8) | 40.5 (+3.8) | 58.7 (+8.2) |\
+| + VL-PRM-7B | 67.6 (+1.6) | 66.8 (+20.6) | 44.2 (+17.3) | 78.3 (+1.4) | 40.1 (+3.2) | 59.4 (+8.9) |\
 ### Gemma-3 Family
 | Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |
+|-------|------|-----------|---------------|-----------|------------|---------|\
+| **Gemma-3-12B** | 57.6 | 45.0 | 29.1 | 58.9 | 28.1 | 43.7 |\
+| + VL-PRM-3B | 60.4 (+2.8) | 57.7 (+12.7) | 39.7 (+10.6) | 60.3 (+1.4) | 33.8 (+5.7) | 50.4 (+6.7) |\
+| + VL-PRM-7B | 60.2 (+2.6) | 59.0 (+12.0) | 41.1 (+4.5) | 63.3 (+4.4) | 33.9 (+5.8) | 51.5 (+7.8) |\
+| **Gemma-3-27B** | 62.9 | 50.8 | 29.9 | 61.6 | 32.4 | 47.5 |\
+| + VL-PRM-3B | 65.5 (+2.6) | 67.4 (+16.6) | 40.3 (+10.4) | 65.4 (+3.8) | 39.8 (+7.4) | 55.7 (+8.2) |\
 | + VL-PRM-7B | 64.5 (+1.6) | 67.6 (+16.8) | 41.1 (+11.2) | 65.2 (+3.6) | 40.9 (+8.5) | 55.9 (+8.4) |
 ### Framework versions