## 1. Model Introduction
Skywork-R1V2-38B is a **state-of-the-art open-source multimodal reasoning model**, achieving top-tier performance across multiple benchmarks:
- On **MMMU**, it scores **73.6%**, the **highest among all open-source models** to date.
- On **OlympiadBench**, it achieves **62.6%**, leading **by a large margin** over other open models.
- R1V2 also performs strongly on **MathVision**, **MMMU-Pro**, and **MathVista**, **rivaling proprietary commercial models**.
- Overall, R1V2 stands out as a **high-performing, open-source VLM** combining powerful **visual reasoning** and **text understanding**.
### 🔧 Model Details
---
## 2. Evaluation
Comparison with Larger-Scale Open-Source ModelsComparison with Proprietary Models
Evaluation Results of State-of-the-Art LLMs and VLMs
Model
Supports Vision
Text Reasoning (%)
Multimodal Reasoning (%)
AIME24
LiveCodebench
liveBench
IFEVAL
BFCL
MATH‑500
AIME 2024
GPQA
MMMU(val)
MathVista(mini)
MathVision(mini)
OlympiadBench
mmmu‑pro
%
%
%
%
%
pass@1
pass@1
pass@1
%
%
%
%
%
R1V2‑38B
✅
78.9
63.6
73.2
82.9
66.3
94.0
72.0
61.6
73.6
74.0
49.0
62.6
52.0
R1V1‑38B
✅
72.0
57.2
54.6
72.5
53.5
–
–
–
68.0
67.0
–
40.4
–
Deepseek‑R1‑671B
❌
74.3
65.9
71.6
83.3
60.3
97.3
79.8
71.5
–
–
–
–
–
GPT‑o1
❌
79.8
63.4
72.2
–
–
–
–
–
–
–
–
–
–
GPT‑o4‑mini
✅
93.4
74.6
78.1
–
–
74.6
9.3
49.9
81.6
84.3
58.0
–
–
Claude 3.5 Sonnet
✅
–
–
–
–
–
78.3
16.0
65.0
66.4
65.3
–
–
–
Kimi k1.5 long-cot
✅
–
–
–
–
–
96.2
77.5
–
70.0
74.9
–
–
–
Qwen2.5‑VL‑72B‑Instruct
✅
–
–
–
–
–
–
–
–
70.2
74.8
–
–
–
InternVL2.5‑78B
✅
–
–
–
–
–
–
–
–
70.1
72.3
–
33.2
–
---
## 3. Usage
### 1. Clone the Repository
```shell
git clone https://github.com/SkyworkAI/Skywork-R1V.git
cd skywork-r1v/inference
```
### 2. Set Up the Environment
```shell
conda create -n r1v-vllm python=3.12
conda activate r1v-vllm
pip install -U vllm
```
### 3. Run the Inference Script
transformers inference
```shell
CUDA_VISIBLE_DEVICES="0,1" python inference_with_transformers.py \
--model_path path \
--image_paths image1_path \
--question "your question"
```
vllm inference
```shell
python inference_with_vllm.py \
--model_path path \
--image_paths image1_path image2_path \
--question "your question" \
--tensor_parallel_size 4
```
---
## 4. Citation
If you use Skywork-R1V in your research, please cite:
```
@misc{chris2025skyworkr1v2,
title = {Skywork R1V2: Visual Hybrid Reinforcement Learning for Multimodal Reasoning},
author = {Chris and Yichen Wei and Yi Peng and Xiaokun Wang and Weijie Qiu and Wei Shen and Tianyidan Xie and Jiangbo Pei and Jianhao Zhang and Yunzhuo Hao and Xuchen Song and Yang Liu and Yahui Zhou},
year = {2025},
institution = {Skywork AI, Kunlun Inc.},
howpublished = {\url{https://github.com/SkyworkAI/Skywork-R1V/blob/main/Skywork_R1V2.pdf}},
}
```
```
@misc{peng2025skyworkr1vpioneeringmultimodal,
title={Skywork R1V: Pioneering Multimodal Reasoning with Chain-of-Thought},
author={Yi Peng and Chris and Xiaokun Wang and Yichen Wei and Jiangbo Pei and Weijie Qiu and Ai Jian and Yunzhuo Hao and Jiachun Pan and Tianyidan Xie and Li Ge and Rongxian Zhuang and Xuchen Song and Yang Liu and Yahui Zhou},
year={2025},
eprint={2504.05599},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2504.05599},
}
```
*This project is released under an open-source license.*
