---
base_model: Qwen/Qwen2.5-VL-7B-Instruct
library_name: transformers
model_name: ob11/Qwen-VL-PRM-7B
licence: apache-2.0
datasets:
- ob11/VL-PRM300K-V1-train
---

# Model Summary

> Qwen-VL-PRM-7B is a process reward model finetuned from Qwen2.5-7B-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/pj4oc0qh
- **Repository:** https://github.com/theogbrand/vlprm
- **Paper:** https://arxiv.org/pdf/2509.23250

# Use

The model usage is documented [here](https://github.com/theogbrand/vlprm/blob/main/eval/tts_eval/reward_guided_search/VisualPRMv2.py).
# Evaluation
### Commercial Models
| Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |
|-------|------|-----------|---------------|-----------|------------|---------|
| GPT-4o | 70.7 | 60.0 | 57.8 | 30.9 | 31.2 | 50.1 |
| o1 | 78.2 | 78.9 | 54.4 | 73.9 | 60.3 | 69.1 |
| 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

- TRL: 0.19.1
- Transformers: 4.55.3
- Pytorch: 2.7.1
- Datasets: 3.0.1
- Tokenizers: 0.21.4

## Citations
    
```bibtex
@misc{ong2025vlprms,
      title={Training Vision-Language Process Reward Models for Test-Time Scaling in Multimodal Reasoning: Key Insights and Lessons Learned}, 
      author={Brandon Ong, Tej Deep Pala, Vernon Toh, William Chandra Tjhi, and Soujanya Poria},
      year={2025},
      eprint={2509.23250},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/pdf/2509.23250}, 
}
```