init: MetaX vllm adapt-task batch submit strategy (zkl)
This commit is contained in:
2
.gitignore
vendored
Normal file
2
.gitignore
vendored
Normal file
@@ -0,0 +1,2 @@
|
||||
.DS_Store
|
||||
__pycache__/
|
||||
15
Dockerfile
Normal file
15
Dockerfile
Normal file
@@ -0,0 +1,15 @@
|
||||
FROM modelhubxc-4pd.tencentcloudcr.com/xc_agent_platform/python:3.11-slim
|
||||
|
||||
ENV PYTHONUNBUFFERED=1
|
||||
|
||||
|
||||
WORKDIR /app
|
||||
|
||||
COPY requirements.txt .
|
||||
RUN pip install --no-cache-dir -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt
|
||||
|
||||
COPY . .
|
||||
|
||||
EXPOSE 8080
|
||||
|
||||
CMD ["python", "main.py"]
|
||||
18
README.md
Normal file
18
README.md
Normal file
@@ -0,0 +1,18 @@
|
||||
# xc_validation_strategy_vllm_gguf_metax_zkl
|
||||
|
||||
批量向 ModelHub XC 平台提交模型适配任务的策略服务(zhouyuanxi 账号,jiajing xc-Token),
|
||||
之后保持 HTTP 服务存活供平台探活。
|
||||
|
||||
## 功能
|
||||
|
||||
- 通过 `/api/adapt/task/add` 接口(xc-Token 认证)批量提交模型适配任务
|
||||
- vllm 框架(api chat),MetaX_c-500 GPU
|
||||
- 提交成功的模型写入 `submitted_adapt_tasks.txt`
|
||||
- 暴露 `/health` 和 `/status` 接口满足平台运行时契约
|
||||
|
||||
## 平台契约说明
|
||||
|
||||
- Dockerfile 位于仓库根目录,基于官方轻量基础镜像
|
||||
- 暴露 8080 端口并实现 `GET /health`
|
||||
- 通过环境变量 `STRATEGY_ID` 获取策略 ID
|
||||
- 正确处理 `SIGTERM` 信号,支持优雅停机
|
||||
1
requirements.txt
Normal file
1
requirements.txt
Normal file
@@ -0,0 +1 @@
|
||||
requests
|
||||
Reference in New Issue
Block a user