Instructions to use OpenGVLab/VisualPRM-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenGVLab/VisualPRM-8B with Transformers:

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

pipe = pipeline("image-text-to-text", model="OpenGVLab/VisualPRM-8B", 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("OpenGVLab/VisualPRM-8B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use OpenGVLab/VisualPRM-8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenGVLab/VisualPRM-8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenGVLab/VisualPRM-8B",
		"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/OpenGVLab/VisualPRM-8B

SGLang

How to use OpenGVLab/VisualPRM-8B 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 "OpenGVLab/VisualPRM-8B" \
    --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": "OpenGVLab/VisualPRM-8B",
		"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 "OpenGVLab/VisualPRM-8B" \
        --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": "OpenGVLab/VisualPRM-8B",
		"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 OpenGVLab/VisualPRM-8B with Docker Model Runner:
```
docker model run hf.co/OpenGVLab/VisualPRM-8B
```

Add paper abstract

by nielsr HF Staff - opened Mar 15, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+15

-12

Files changed (1) hide show

README.md +15 -12

README.md CHANGED Viewed

@@ -1,19 +1,19 @@
 ---
-license: mit
-pipeline_tag: image-text-to-text
-library_name: transformers
 base_model:
-  - OpenGVLab/InternVL2_5-8B
-  - OpenGVLab/InternVL2_5-8B-MPO
-base_model_relation: finetune
 datasets:
-  - OpenGVLab/MMPR-v1.1
-  - OpenGVLab/VisualPRM400K
 language:
-  - multilingual
 tags:
-  - internvl
-  - custom_code
 ---
 # VisualPRM-8B
@@ -36,7 +36,6 @@ We introduce VisualPRM, an advanced multimodal Process Reward Model (PRM) with 8
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/619507e7b74b6c591f794340/XVv8M4u-GOOJknrSnGWI4.png)
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/619507e7b74b6c591f794340/DjMtYbrZo0J-LPVk_qiCt.png)
 ## Inference with Transformers
 ```python
@@ -205,3 +204,7 @@ If you find this project useful in your research, please consider citing:
   year={2024}
 }
 ```

 ---
 base_model:
+- OpenGVLab/InternVL2_5-8B
+- OpenGVLab/InternVL2_5-8B-MPO
 datasets:
+- OpenGVLab/MMPR-v1.1
+- OpenGVLab/VisualPRM400K
 language:
+- multilingual
+library_name: transformers
+license: mit
+pipeline_tag: image-text-to-text
 tags:
+- internvl
+- custom_code
+base_model_relation: finetune
 ---
 # VisualPRM-8B
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/619507e7b74b6c591f794340/XVv8M4u-GOOJknrSnGWI4.png)
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/619507e7b74b6c591f794340/DjMtYbrZo0J-LPVk_qiCt.png)
 ## Inference with Transformers
 ```python
   year={2024}
 }
 ```
+## Abstract
+We introduce VisualPRM, an advanced multimodal Process Reward Model (PRM) with 8B parameters, which improves the reasoning abilities of existing Multimodal Large Language Models (MLLMs) across different model scales and families with Best-of-N (BoN) evaluation strategies. **Specifically, our model improves the reasoning performance of three types of MLLMs and four different model scales. Even when applied to the highly capable InternVL2.5-78B, it achieves a 5.9-point improvement across seven multimodal reasoning benchmarks.** Experimental results show that our model exhibits superior performance compared to Outcome Reward Models and Self-Consistency during BoN evaluation. To facilitate the training of multimodal PRMs, we construct a multimodal process supervision dataset VisualPRM400K using an automated data pipeline. For the evaluation of multimodal PRMs, we propose VisualProcessBench, a benchmark with human-annotated step-wise correctness labels, to measure the abilities of PRMs to detect erroneous steps in multimodal reasoning tasks. We hope that our work can inspire more future research and contribute to the development of MLLMs.