A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 16, 19:30 UTC
[CI/Build] bump minimum cmake version (#6999)
Daniele committed
Aug 13, 2024 at 01:29 UTC
774cd1d3bf7890c6abae6c7ace798c4a376b2b20
4 files changed
+3
-5
CMakeLists.txt
+1
-1
@@ -1,4 +1,4 @@
1
-cmake_minimum_required(VERSION 3.21)
1
+cmake_minimum_required(VERSION 3.26)
2
3
project(vllm_extensions LANGUAGES CXX)
4
pyproject.toml
+1
-1
@@ -1,7 +1,7 @@
1
[build-system]
2
# Should be mirrored in requirements-build.txt
3
requires = [
4
- "cmake>=3.21",
4
+ "cmake>=3.26",
5
"ninja",
6
"packaging",
7
"setuptools >= 49.4.0",
requirements-build.txt
+1
-1
@@ -1,5 +1,5 @@
1
# Should be mirrored in pyproject.toml
2
-cmake>=3.21
2
+cmake>=3.26
3
ninja
4
packaging
5
setuptools>=49.4.0
requirements-common.txt
-2
@@ -1,5 +1,3 @@
1
-cmake >= 3.21
2
-ninja # For faster builds.
1
psutil
2
sentencepiece # Required for LLaMA tokenizer.
3
numpy < 2.0.0