Add library_name, license and pipeline_tag
Browse filesThis PR adds the `pipeline_tag`, `library_name`, and `license` to the model card metadata. This improves discoverability and clarity. This PR ensures people can find your model at https://huggingface.co/models?pipeline_tag=text-generation and will also populate the "how to use" button.
README.md
CHANGED
|
@@ -3,11 +3,19 @@ base_model:
|
|
| 3 |
- deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
|
| 4 |
datasets:
|
| 5 |
- Skywork/Skywork-OR1-RL-Data
|
|
|
|
|
|
|
|
|
|
| 6 |
---
|
|
|
|
| 7 |
<div align="center">
|
| 8 |
|
| 9 |
# 🤔 Skywork-OR1 (Open Reasoner 1)
|
| 10 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
</div>
|
| 12 |
<div>
|
| 13 |
<br>
|
|
@@ -64,12 +72,12 @@ We include the detailed results in the following table.
|
|
| 64 |
|-------|---------|---------|--------------|
|
| 65 |
| DeepSeek-R1-Distill-Qwen-7B | 55.5 | 39.2| 37.6 |
|
| 66 |
| Light-R1-7B-DS | 59.1 | 44.3| 39.5 |
|
|
|
|
|
|
|
| 67 |
| DeepSeek-R1-Distill-Qwen-32B | 72.9 | 59.0| 57.2 |
|
| 68 |
| TinyR1-32B-Preview | 78.1| 65.3| 61.6 |
|
| 69 |
| QwQ-32B | 79.5 | 65.3| 61.6 |
|
| 70 |
| DeepSeek-R1 | 79.8 | 70.0| 65.9 |
|
| 71 |
-
| **Skywork-OR1-Math-7B** | 69.8 | 52.3 | 43.6 |
|
| 72 |
-
| **Skywork-OR1-7B-Preview** | 63.6 | 45.8 | 43.9 |
|
| 73 |
| **Skywork-OR1-32B-Preview** | 79.7 | 69.0 | 63.9 |
|
| 74 |
|
| 75 |
## ⚙️ Training Recipe
|
|
@@ -111,7 +119,7 @@ Please cite the following:
|
|
| 111 |
|
| 112 |
@misc{skywork-or1-2025,
|
| 113 |
title={Skywork Open Reasoner Series},
|
| 114 |
-
author = {He, Jujie and Liu, Jiacai and Liu, Chris Yuhao and Yan, Rui and Wang, Chaojie and Cheng, Peng and Zhang, Xiaoyu and Zhang, Fuxiang and Xu, Jiacheng and Shen, Wei and Li, Siyuan and Zeng, Liang and Wei, Tianwen and Cheng, Cheng and Liu, Yang and Zhou, Yahui},
|
| 115 |
howpublished={\url{https://capricious-hydrogen-41c.notion.site/Skywork-Open-Reaonser-Series-1d0bc9ae823a80459b46c149e4f51680}},
|
| 116 |
note={Notion Blog},
|
| 117 |
year={2025}
|
|
|
|
| 3 |
- deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
|
| 4 |
datasets:
|
| 5 |
- Skywork/Skywork-OR1-RL-Data
|
| 6 |
+
library_name: transformers
|
| 7 |
+
license: apache-2.0
|
| 8 |
+
pipeline_tag: text-generation
|
| 9 |
---
|
| 10 |
+
|
| 11 |
<div align="center">
|
| 12 |
|
| 13 |
# 🤔 Skywork-OR1 (Open Reasoner 1)
|
| 14 |
|
| 15 |
+
<div>
|
| 16 |
+
✊ Unleashing the Power of Reinforcement Learning for Math and Code Reasoners 🤖
|
| 17 |
+
</div>
|
| 18 |
+
|
| 19 |
</div>
|
| 20 |
<div>
|
| 21 |
<br>
|
|
|
|
| 72 |
|-------|---------|---------|--------------|
|
| 73 |
| DeepSeek-R1-Distill-Qwen-7B | 55.5 | 39.2| 37.6 |
|
| 74 |
| Light-R1-7B-DS | 59.1 | 44.3| 39.5 |
|
| 75 |
+
| **Skywork-OR1-Math-7B** | 69.8 | 52.3 | 43.6 |
|
| 76 |
+
| **Skywork-OR1-7B-Preview** | 63.6 | 45.8 | 43.9 |
|
| 77 |
| DeepSeek-R1-Distill-Qwen-32B | 72.9 | 59.0| 57.2 |
|
| 78 |
| TinyR1-32B-Preview | 78.1| 65.3| 61.6 |
|
| 79 |
| QwQ-32B | 79.5 | 65.3| 61.6 |
|
| 80 |
| DeepSeek-R1 | 79.8 | 70.0| 65.9 |
|
|
|
|
|
|
|
| 81 |
| **Skywork-OR1-32B-Preview** | 79.7 | 69.0 | 63.9 |
|
| 82 |
|
| 83 |
## ⚙️ Training Recipe
|
|
|
|
| 119 |
|
| 120 |
@misc{skywork-or1-2025,
|
| 121 |
title={Skywork Open Reasoner Series},
|
| 122 |
+
author = {He, Jujie and Liu, Jiacai and Liu, Chris Yuhao and Yan, Rui and Wang, Chaojie and Cheng, Peng and Zhang, Xiaoyu and Zhang, Fuxiang and Xu, Jiacheng and Shen, Wei and Li, Siyuan and Zeng, Liang and Wei, Tianwen and Cheng, Cheng and An, Bo and Liu, Yang and Zhou, Yahui},
|
| 123 |
howpublished={\url{https://capricious-hydrogen-41c.notion.site/Skywork-Open-Reaonser-Series-1d0bc9ae823a80459b46c149e4f51680}},
|
| 124 |
note={Notion Blog},
|
| 125 |
year={2025}
|