Commit 5d6aae4d authored by zzg_666's avatar zzg_666
Browse files

first commit

parents
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
# MiroThinker-v1.5
## 论文
[MiroThinker-v1.5](https://arxiv.org/abs/2511.11793)
## 模型简介
MiroThinkerv1.5是全球领先的搜索智能体,旨在提升工具增强推理与信息获取能力。与以往仅扩展模型规模或上下文长度的智能体不同,MiroThinker在模型层面引入了交互式扩展机制,将模型系统性训练为能够处理更深层次、更频繁的智能体‑环境交互,以此作为性能提升的第三维度。交互式扩展借助环境反馈与外部信息获取,实现错误修正与执行路径优化。实证结果验证了交互式扩展的有效性:在多项基准测试中,随着模型与环境的交互深度与频率逐步提升,其性能呈现出可预测的持续进步。
## 环境依赖
| 软件 | 版本 |
| :------: | :------: |
| DTK | 25.04.2 |
| python | 3.10.12 |
| transformers | >=4.57.1 |
| vllm | 0.11.0+das.opt1.alpha.dtk25042.20251225.gca4598a4 |
| torch | 2.5.1+das.opt1.dtk25042 |
| triton | 3.1.0+das.opt1.dtk25042.20251224.gaa867475 |
推荐使用镜像:harbor.sourcefind.cn:5443/dcu/admin/base/vllm:0.11.0-ubuntu22.04-dtk25.04.2-tx-1226-das1.7-py3.10-20251226
- 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 MiroThinker --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/ harbor.sourcefind.cn:5443/dcu/admin/base/vllm:0.11.0-ubuntu22.04-dtk25.04.2-tx-1226-das1.7-py3.10-20251226 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 miromind-ai/MiroThinker-v1.5-235B --trust-remote-code --dtype float16 -tp 8 --max-model-len 32768 --gpu-memory-utilization 0.95 --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": "miromind-ai/MiroThinker-v1.5-235B",
"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|
|:-----:|:----------:|:----------:|:---------------------:|:----------:|
| MiroThinker-v1.5-235B | 235B | BW1000 | 8 | [下载地址](https://huggingface.co/miromind-ai/MiroThinker-v1.5-235B) |
## 源码仓库及问题反馈
- https://developer.sourcefind.cn/codes/modelzoo/mirothinker-v1.5_vllm
## 参考资料
- https://huggingface.co/miromind-ai/MiroThinker-v1.5-235B
icon.png

62.1 KB

# 模型唯一标识
modelCode=1946
# 模型名称
modelName=MiroThinker-v1.5_vllm
# 模型描述
modelDescription=MiroThinkerv1.5是全球领先的搜索智能体,旨在提升工具增强推理与信息获取能力
# 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=BW1000
curl -X POST "http://localhost:8010/v1/chat/completions" -H "Content-Type: application/json" --data '{
"model": "miromind-ai/MiroThinker-v1.5-235B",
"messages": [
{
"role": "user",
"content": "请介绍下自己"
}
]
}'
\ No newline at end of file
vllm serve miromind-ai/MiroThinker-v1.5-235B --trust-remote-code --dtype float16 -tp 8 --max-model-len 32768 --gpu-memory-utilization 0.95 --port 8010
\ No newline at end of file
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment