Skip to content

Commit 2335fe7

Browse files
committed
bump version to v0.13.0
1 parent 8d7980b commit 2335fe7

6 files changed

Lines changed: 11 additions & 145 deletions

File tree

.github/workflows/cuda12.8_whl_release.yml

Lines changed: 0 additions & 121 deletions
This file was deleted.

.github/workflows/pypi.yml

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -18,8 +18,8 @@ jobs:
1818
env:
1919
PYTHON_VERSION: ${{ matrix.pyver }}
2020
PLAT_NAME: manylinux2014_x86_64
21-
DOCKER_TAG: cuda12.4
22-
OUTPUT_FOLDER: cuda12_dist
21+
DOCKER_TAG: cuda12.8
22+
OUTPUT_FOLDER: cuda12.8_dist
2323
steps:
2424
- name: Free disk space
2525
uses: jlumbroso/free-disk-space@main
@@ -75,11 +75,11 @@ jobs:
7575
shell: pwsh
7676
run: ./builder/windows/setup_cuda.ps1
7777
env:
78-
INPUT_CUDA_VERSION: '12.6.2'
78+
INPUT_CUDA_VERSION: '12.8.1'
7979
- name: Build wheel
8080
run: |
8181
python -m build --wheel -o build/wheel
82-
Get-ChildItem -Path "build" -Filter "*.whl" | ForEach-Object { change_wheel_version $_.FullName --local-version cu121 --delete-old-wheel }
82+
Get-ChildItem -Path "build" -Filter "*.whl" | ForEach-Object { change_wheel_version $_.FullName --local-version cu128 --delete-old-wheel }
8383
- name: Upload Artifacts
8484
uses: actions/upload-artifact@v4
8585
with:

README.md

Lines changed: 1 addition & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -224,15 +224,7 @@ conda activate lmdeploy
224224
pip install lmdeploy
225225
```
226226

227-
Since v0.3.0, the default prebuilt package is compiled on **CUDA 12**. Starting from v0.10.2, LMDeploy no longer supports CUDA 11 series.
228-
229-
If you are using a GeForce RTX 50 series graphics card, please install the LMDeploy prebuilt package compiled with **CUDA 12.8** as follows:
230-
231-
```shell
232-
export LMDEPLOY_VERSION=0.12.3
233-
export PYTHON_VERSION=312
234-
pip install https://github.com/InternLM/lmdeploy/releases/download/v${LMDEPLOY_VERSION}/lmdeploy-${LMDEPLOY_VERSION}+cu128-cp${PYTHON_VERSION}-cp${PYTHON_VERSION}-manylinux2014_x86_64.whl --extra-index-url https://download.pytorch.org/whl/cu128
235-
```
227+
Starting from **v0.13.0**, the default prebuilt wheels published on **PyPI** are built against **CUDA 12.8**, so `pip install lmdeploy` is sufficient for typical setups including GeForce RTX 50 series.
236228

237229
## Offline Batch Inference
238230

README_ja.md

Lines changed: 4 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -201,7 +201,10 @@ conda activate lmdeploy
201201
pip install lmdeploy
202202
```
203203

204-
v0.3.0から、デフォルトの事前構築済みパッケージはCUDA 12でコンパイルされています。
204+
**v0.13.0** 以降、**PyPI** に公開される既定の事前構築wheelは **CUDA 12.8** 向けにビルドされています。v0.10.2以降、LMDeployはCUDA 11系をサポートしません。
205+
206+
GeForce RTX 50シリーズを含む一般的な用途でも、上記の `pip install lmdeploy` で問題ありません。
207+
205208
CUDA 11+プラットフォームでのインストールに関する情報、またはソースからのビルド手順については、[インストールガイドを](docs/en/get_started/installation.md)参照してください。
206209

207210
## オフラインバッチ推論

README_zh-CN.md

Lines changed: 1 addition & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -226,15 +226,7 @@ conda activate lmdeploy
226226
pip install lmdeploy
227227
```
228228

229-
自 v0.3.0 版本起,默认预编译包基于 **CUDA 12** 编译。v0.10.2 及更高版本中,LMDeploy 不再支持 CUDA 11+。
230-
231-
若使用 GeForce RTX 50 系列显卡,请按照如下方式安装基于 **CUDA 12.8** 编译的 LMDeploy 预编译包。
232-
233-
```shell
234-
export LMDEPLOY_VERSION=0.12.3
235-
export PYTHON_VERSION=312
236-
pip install https://github.com/InternLM/lmdeploy/releases/download/v${LMDEPLOY_VERSION}/lmdeploy-${LMDEPLOY_VERSION}+cu128-cp${PYTHON_VERSION}-cp${PYTHON_VERSION}-manylinux2014_x86_64.whl --extra-index-url https://download.pytorch.org/whl/cu128
237-
```
229+
**v0.13.0** 起,**PyPI** 上默认预编译 wheel 基于 **CUDA 12.8** 构建,一般用户(含 GeForce RTX 50 系列)使用上方的 `pip install lmdeploy` 即可。
238230

239231
## 离线批处理
240232

lmdeploy/version.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
# Copyright (c) OpenMMLab. All rights reserved.
22

3-
__version__ = '0.12.3'
3+
__version__ = '0.13.0'
44
short_version = __version__
55

66

0 commit comments

Comments
 (0)