first commit

LICENSE

+MIT License
+Copyright (c) [year] [fullname]
+Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
+The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
\ No newline at end of file

README.md

+# PaCoRe
+## 论文
+[PaCoRe](https://github.com/stepfun-ai/PaCoRe/blob/main/pacore_report.pdf)
+
+## 模型简介
+PaCoRe（Parallel Coordinated Reasoning，并行协同推理）是一种将推理驱动力从顺序深度转向协同并行广度的框架，突破了模型上下文限制，并实现测试阶段计算资源的规模化扩展：  
+- 并行思维探索：PaCoRe启动大规模并行探索路径；  
+- 多轮协同整合：通过消息传递架构将并行思考压缩为精炼信息，并综合指导下一轮推理；  
+- 基于结果的强化学习训练：通过大规模结果导向的强化学习训练，PaCoRe掌握了调和多元并行洞察所需的推理综合能力。  
+该方法在多个领域实现显著性能提升，尤其在数学推理方面超越前沿系统：通过将有效测试时计算量扩展至约200万token，一个80亿参数的模型在HMMT2025基准测试中达到94.5%的准确率，超越GPT-5的93.2%。
+<div align=center>
+    <img src="./doc/perf.png"/>
+</div>
+## 环境依赖
+
+| 软件 | 版本 |
+| :------: | :------: |
+| DTK | 25.04.2 |
+| python | 3.10.12 |
+| transformers | >=4.57.1 |
+| vllm |  0.9.2+das.opt1.dtk25042 |
+| torch | 2.5.1+das.opt1.dtk25042 |
+| triton | 3.1+das.opt1.3c5d12d.dtk25041 |
+| flash_attn | 2.6.1+das.opt1.dtk2504 |
+| flash_mla | 1.0.0+das.opt1.dtk25042 |
+
+推荐使用镜像:image.sourcefind.cn:5000/dcu/admin/base/vllm:0.9.2-ubuntu22.04-dtk25.04.2-py3.10
+- 挂载地址`-v`根据实际模型情况修改
+
+```bash
+docker run -it --shm-size 60g --network=host --name PaCoRe --privileged --device=/dev/kfd --device=/dev/dri --device=/dev/mkfd --group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined -u root -v /opt/hyhal/:/opt/hyhal/:ro -v /path/your_code_path/:/path/your_code_path/  image.sourcefind.cn:5000/dcu/admin/base/vllm:0.9.2-ubuntu22.04-dtk25.04.2-py3.10 bash
+```
+更多镜像可前往[光源](https://sourcefind.cn/#/service-list)下载使用。
+
+关于本项目DCU显卡所需的特殊深度学习库可从[光合](https://developer.sourcefind.cn/tool/)开发者社区下载安装。
+
+## 数据集
+暂无
+
+## 训练
+暂无
+
+## 推理
+
+### vllm
+#### 单机推理
+
+```bash
+## serve启动
+可参考vllm_serve.sh脚本
+vllm serve stepfun-ai/PaCoRe-8B  --trust-remote-code --dtype bfloat16 --max-seq-len-to-capture 32768 -tp 1 --max-model-len 32768 --port 8010
+## client访问
+可参考vllm_cilent.sh
+curl -X POST "http://localhost:8010/v1/chat/completions"      -H "Content-Type: application/json"     --data '{
+                "model": "stepfun-ai/PaCoRe-8B",
+                "messages": [
+                        {
+                                "role": "user",
+                                "content": "请介绍下自己"
+                        }
+                ]
+        }'
+
+```
+
+## 效果展示
+<div align=center>
+    <img src="./doc/result.png"/>
+</div>
+
+### 精度
+DCU与GPU精度一致，推理框架：vllm。
+
+## 预训练权重
+| 模型名称  | 权重大小  | DCU型号  | 最低卡数需求 |下载地址|
+|:-----:|:----------:|:----------:|:---------------------:|:----------:|
+| PaCoRe | 8B | K100AI | 1 | [下载地址](https://huggingface.co/stepfun-ai/PaCoRe-8B) |
+
+## 源码仓库及问题反馈
+- https://developer.sourcefind.cn/codes/modelzoo/pacore_vllm
+
+## 参考资料
+- https://github.com/stepfun-ai/PaCoRe/tree/main

model.properties

+# 模型唯一标识
+modelCode=1924
+# 模型名称
+modelName=PaCoRe_vllm
+# 模型描述
+modelDescription=PaCoRe（Parallel Coordinated Reasoning，并行协同推理）是一种将推理驱动力从顺序深度转向协同并行广度的框架，突破了模型上下文限制，并实现测试阶段计算资源的规模化扩展
+# 应用场景
+processType=推理
+# 算法类别
+appScenario=文本生成
+# 框架类型
+frameType=vllm
+# 加速卡类型
+accelerateType=K100AI

vllm_cilent.sh

+curl -X POST "http://localhost:8010/v1/chat/completions"      -H "Content-Type: application/json"     --data '{
+                "model": "stepfun-ai/PaCoRe-8B",
+                "messages": [
+                        {
+                                "role": "user",
+                                "content": "请介绍下自己"
+                        }
+                ]
+        }'

vllm_serve.sh

+vllm serve stepfun-ai/PaCoRe-8B  --trust-remote-code --dtype bfloat16 --max-seq-len-to-capture 32768 -tp 1 --max-model-len 32768 --port 8010
\ No newline at end of file