Update README.md

README.md

@@ -9,7 +9,7 @@ pipeline_tag: image-text-to-text
   <img src="skywork-logo.png" alt="Introduction Image" width="500" height="400"> 
 </div>
 
-## 📖 [R1V2 Report](https://arxiv.org/abs/2504.16656) | 💻 [GitHub](https://github.com/SkyworkAI/Skywork-R1V) | 🌐 [ModelScope](https://modelscope.cn/models/Skywork/Skywork-R1V2-38B) 
+## 📖 [R1V3 Report](https://arxiv.org/abs/2507.06167) | 💻 [GitHub](https://github.com/SkyworkAI/Skywork-R1V) | 🌐 [ModelScope](https://modelscope.cn/models/Skywork/Skywork-R1V3-38B) 
 
 <div align="center">
 
@@ -29,151 +29,72 @@ pipeline_tag: image-text-to-text
       <th>Model</th>
       <th align="center">MMMU</th>
       <th align="center">MathVista</th>
-      <th align="center">MathVision</th>
-      <th align="center">Olympiad Bench</th>
-      <th align="center">AIME 24</th>
-      <th align="center">LiveCode bench</th>
-      <th align="center">Live Bench</th>
-      <th align="center">IFEVAL</th>
     </tr>
   </thead>
   <tbody>
     <tr>
-      <td colspan="9" align="center"><i>Proprietary Models</i></td>
+      <td colspan="3" align="center"><i>Proprietary Models</i></td>
     </tr>
     <tr>
       <td>Claude-3.5-Sonnet</td>
       <td align="center">70.4</td>
-      <td align="center">67.7</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
       <td align="center">-</td>
     </tr>
     <tr>
       <td>Gemini-2-Flash</td>
       <td align="center">70.7</td>
-      <td align="center">73.1</td>
-      <td align="center">41.3</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
       <td align="center">-</td>
     </tr>
     <tr>
       <td>Kimi-k1.5-longcot</td>
       <td align="center">70.0</td>
-      <td align="center">74.9</td>
-      <td align="center">53.3</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
       <td align="center">-</td>
     </tr>
     <tr>
       <td>OpenAI-o1</td>
       <td align="center">-</td>
       <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">74.3</td>
-      <td align="center">63.4</td>
-      <td align="center">72.2</td>
-      <td align="center">-</td>
     </tr>
     <tr>
       <td>OpenAI-o4-mini</td>
       <td align="center"><b>81.6</b></td>
-      <td align="center"><b>84.3</b></td>
-      <td align="center"><b>58.0</b></td>
-      <td align="center">-</td>
-      <td align="center"><b>93.4</b></td>
-      <td align="center"><b>74.6</b></td>
-      <td align="center"><b>78.1</b></td>
       <td align="center">-</td>
     </tr>
     <tr>
-      <td colspan="9" align="center"><i>Open-Source Models</i></td>
+      <td colspan="3" align="center"><i>Open-Source Models</i></td>
     </tr>
     <tr>
       <td>Skywork-R1V1</td>
       <td align="center">68.0</td>
-      <td align="center">67.0</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">72.0</td>
-      <td align="center">57.2</td>
-      <td align="center">54.6</td>
       <td align="center">72.5</td>
     </tr>
     <tr>
       <td>DeepseekR1-671B</td>
       <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td
->
-      <td align="center"><b>79.8</b></td>
-      <td align="center"><b>65.9</b></td>
-      <td align="center">71.6</td>
       <td align="center"><b>83.3</b></td>
     </tr>
     <tr>
       <td>InternVL3-38B</td>
       <td align="center">70.1</td>
-      <td align="center">75.1</td>
-      <td align="center">34.2</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
       <td align="center">-</td>
     </tr>
     <tr>
       <td>Qwen2.5-VL-72B</td>
       <td align="center">70.2</td>
-      <td align="center">74.8</td>
-      <td align="center">38.1</td>
-      <td align="center">40.4</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
       <td align="center">-</td>
     </tr>
     <tr>
       <td>QvQ-Preview-72B</td>
       <td align="center">70.3</td>
-      <td align="center">71.4</td>
-      <td align="center">35.9</td>
-      <td align="center">33.2</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
-      <td align="center">-</td>
       <td align="center">-</td>
     </tr>
     <tr>
-      <td>Skywork-R1V2</td>
-      <td align="center"><b>73.6</b></td>
-      <td align="center">74.0</td>
-      <td align="center"><b>49.0</b></td>
-      <td align="center"><b>62.6</b></td>
-      <td align="center">78.9</td>
-      <td align="center">63.6</td>
+      <td>Skywork-R1V3</td>
       <td align="center"><b>73.2</b></td>
       <td align="center">82.9</td>
     </tr>
     <tr>
-      <td>Skywork-R1V2-AWQ</td>
-      <td align="center">64.4</td>
-      <td align="center">64.8</td>
-      <td align="center">42.9</td>
-      <td align="center">54.8</td>
-      <td align="center">77.3</td>
-      <td align="center">55.7</td>
+      <td>Skywork-R1V3-AWQ</td>
       <td align="center">64.1</td>
       <td align="center">72.5</td>
     </tr>
@@ -191,7 +112,7 @@ You can use the quantized model with different inference frameworks:
 import os
 from vllm import LLM, SamplingParams
 from vllm.entrypoints.chat_utils import load_chat_template
-model_name = "Skywork/Skywork-R1V2-38B-AWQ"  # or local path
+model_name = "Skywork/Skywork-R1V3-38B-AWQ"  # or local path
 llm = LLM(model_name, 
           dtype='float16', 
           quantization="awq", 
@@ -205,7 +126,7 @@ llm = LLM(model_name,
 #### OpenAI-compatible API Server
 
 ```bash
-MODEL_ID="Skywork/Skywork-R1V2-38B-AWQ"  # or local path
+MODEL_ID="Skywork/Skywork-R1V3-38B-AWQ"  # or local path
 CUDA_VISIBLE_DEVICES=0 \
     python -m vllm.entrypoints.openai.api_server \
     --model $MODEL_ID \
@@ -223,7 +144,7 @@ CUDA_VISIBLE_DEVICES=0 \
 import os
 from lmdeploy import pipeline, TurbomindEngineConfig, ChatTemplateConfig
 from lmdeploy.vl import load_image
-model_path = "Skywork/Skywork-R1V2-38B-AWQ"  # or local path
+model_path = "Skywork/Skywork-R1V3-38B-AWQ"  # or local path
 engine_config = TurbomindEngineConfig(cache_max_entry_count=0.75) 
 chat_template_config = ChatTemplateConfig(model_name=model_path)
 pipe = pipeline(model_path, 
@@ -270,3 +191,15 @@ If you use this model in your research, please cite:
       url={https://arxiv.org/abs/2504.16656}, 
 }
 ```
+
+```bibtex
+@misc{shen2025skyworkr1v3technicalreport,
+      title={Skywork-R1V3 Technical Report}, 
+      author={Wei Shen and Jiangbo Pei and Yi Peng and Xuchen Song and Yang Liu and Jian Peng and Haofeng Sun and Yunzhuo Hao and Peiyu Wang and Jianhao Zhang and Yahui Zhou},
+      year={2025},
+      eprint={2507.06167},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={https://arxiv.org/abs/2507.06167}, 
+}
+```