apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Docs] Fix nvidia_gpu.md, add sm80 in precompiled (#5462)

Browse Source 
* Update supported GPU architectures in installation guide

* Update supported architectures in GPU installation guide

* Update GPU architectures support in installation guide
This commit is contained in:
Echo-Nie
2025-12-11 14:41:50 +08:00
committed by GitHub
parent 9f4512c932
commit e1347be4d9
2 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
docs/get_started/installation/nvidia_gpu.md
View File

@@ -83,7 +83,7 @@ The built packages will be in the ```FastDeploy/dist``` directory.
## 5. Precompiled Operator Wheel Packages
FastDeploy provides precompiled GPU operator wheel packages for quick setup without building the entire source code.
This method currently supports **SM90 architecture (e.g., H20/H100)** and **CUDA 12.6** environments only.
This method currently supports **SM80/90 architecture (e.g., A100/H100)** and **CUDA 12.6** environments only.
> By default, `build.sh` compiles all custom operators from source.To use the precompiled package, enable it with the `FD_USE_PRECOMPILED` parameter.
> If the precompiled package cannot be downloaded or does not match the current environment, the system will automatically fall back to `4. Build Wheel from Source`.
@@ -104,7 +104,7 @@ cd FastDeploy
# Argument 1: Whether to build wheel package (1 for yes)
# Argument 2: Python interpreter path
# Argument 3: Whether to compile CPU inference operators (false for GPU only)
# Argument 4: Target GPU architectures (currently supports [90])
# Argument 4: Target GPU architectures (currently supports 80/90)
# Argument 5: Whether to use precompiled operators (1 for enable)
# Argument 6 (optional): Specific commitID for precompiled operators(The default is the current commit ID.)
@@ -121,7 +121,7 @@ After the build completes, the operator binaries can be found in `FastDeploy/fas
> **Notes:**
>
> - This mode prioritizes downloading precompiled GPU operator wheels to reduce build time.
> - Currently supports **GPU + SM90 + CUDA 12.6** only.
> - Currently supports **GPU, SM80/90, CUDA 12.6** only.
> - For custom architectures or modified operator logic, please use **source compilation (Section 4)**.
> - You can check whether the precompiled wheel for a specific commit has been successfully built on the [FastDeploy CI Build Status Page](https://github.com/PaddlePaddle/FastDeploy/actions/workflows/ci_image_update.yml).

6
docs/zh/get_started/installation/nvidia_gpu.md
View File

@@ -90,7 +90,7 @@ bash build.sh 1 python false [80,90]
## 5. 算子预编译 Wheel 包
FastDeploy 提供了 GPU 算子预编译版 Wheel 包，可在无需完整源码编译的情况下快速构建。该方式当前仅支持 **SM90 架构（H20/H100等）** 和 **CUDA 12.6** 环境。
FastDeploy 提供了 GPU 算子预编译版 Wheel 包，可在无需完整源码编译的情况下快速构建。该方式当前仅支持 **SM80/90 架构（A100/H100等）** 和 **CUDA 12.6** 环境。
>默认情况下，`build.sh` 会从源码编译；若希望使用预编译包，可使用`FD_USE_PRECOMPILED` 参数；
>若预编译包下载失败或与环境不匹配，系统会自动回退至 `4. wheel 包源码编译` 模式。
@@ -110,7 +110,7 @@ cd FastDeploy
# 第1个参数: 是否打包成 wheel (1 表示打包)
# 第2个参数: Python 解释器路径
# 第3个参数: 是否编译 CPU 推理算子 (false 表示仅 GPU)
# 第4个参数: GPU 架构 (当前仅支持 [90])
# 第4个参数: GPU 架构 (当前仅支持 80/90)
# 第5个参数: 是否使用预编译算子 (1 表示启用预编译)
# 第6个参数(可选): 指定预编译算子的 commitID（默认使用当前的 commitID）
@@ -127,7 +127,7 @@ bash build.sh 1 python false [90] 1 8a9e7b53af4a98583cab65e4b44e3265a93e56d2
> **说明：**
> - 该模式会优先下载预编译的 GPU 算子 whl 包，减少编译时间；
> - 目前仅支持 **GPU + SM90 + CUDA 12.6**；
> - 目前仅支持 **GPU， SM80/90 架构， CUDA 12.6**；
> - 若希望自定义架构或修改算子逻辑，请使用 **源码编译方式（第4节）**。
> - 您可以在 FastDeploy CI 构建状态页面查看对应 commit 的预编译 whl 是否已构建成功。