mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-12-24 13:28:13 +08:00
polish code with new pre-commit rule (#2923)
This commit is contained in:
@@ -37,6 +37,7 @@ image.sourcefind.cn:5000/dcu/admin/base/custom:fastdeploy2.0.0-kylinv10-dtk25.04
|
||||
```
|
||||
|
||||
## 2. 启动服务
|
||||
|
||||
```bash
|
||||
export FD_ATTENTION_BACKEND="BLOCK_ATTN"
|
||||
python -m fastdeploy.entrypoints.openai.api_server \
|
||||
@@ -47,7 +48,7 @@ python -m fastdeploy.entrypoints.openai.api_server \
|
||||
--gpu-memory-utilization=0.8
|
||||
```
|
||||
|
||||
#### 请求服务
|
||||
### 请求服务
|
||||
|
||||
您可以基于 OpenAI 协议,通过 curl 和 python 两种方式请求服务。
|
||||
|
||||
@@ -78,4 +79,4 @@ response = client.chat.completions.create(
|
||||
stream=False,
|
||||
)
|
||||
print(response)
|
||||
```
|
||||
```
|
||||
|
||||
Reference in New Issue
Block a user