[CI] Add per-job uv venv isolation and upgrade CI version to Cuda 13 (#23119)

Co-authored-by: Kangyan Zhou <zky314343421@gmail.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Co-authored-by: Alison Shao <a.shao@wustl.edu>
Co-authored-by: Mick <mickjagger19@icloud.com>
This commit is contained in:
Baizhou Zhang
2026-04-19 05:32:36 -07:00
committed by GitHub
co-authored by Kangyan Zhou Claude Opus 4.7 Alison Shao Mick
parent 03828f4205
commit 6ecd6f84db
39 changed files with 892 additions and 239 deletions
+17 -10
View File
@@ -22,7 +22,7 @@ dependencies = [
"blobfile==3.0.0",
"build",
"compressed-tensors",
"cuda-python==12.9",
"cuda-python>=13.0",
"decord2 ; sys_platform == 'linux' and (platform_machine == 'aarch64' or platform_machine == 'arm64' or platform_machine == 'armv7l')",
"datasets",
"einops",
@@ -37,7 +37,7 @@ dependencies = [
"ninja",
"easydict", # Required by remote model code (e.g. DeepSeek-OCR) loaded via trust_remote_code; validated by transformers 5.4+ check_imports
"numpy",
"nvidia-cutlass-dsl>=4.4.1",
"nvidia-cutlass-dsl==4.4.2",
"nvidia-ml-py",
"openai-harmony==0.0.4",
"openai==2.6.1",
@@ -58,14 +58,14 @@ dependencies = [
"scipy",
"sentencepiece",
"setproctitle",
"flash-attn-4>=4.0.0b4",
"flash-attn-4>=4.0.0b9",
"sglang-kernel==0.4.1",
"soundfile==0.13.1",
"tiktoken",
"timm==1.0.16",
"torch_memory_saver==0.0.9",
"torch==2.9.1",
"torchao==0.9.0",
"torchao==0.17.0",
"torchaudio==2.9.1",
"torchcodec==0.9.1 ; sys_platform != 'linux' or (sys_platform == 'linux' and platform_machine != 'aarch64' and platform_machine != 'arm64' and platform_machine != 'armv7l')", # torchcodec 0.9.1 for torch 2.9.x. Not available on Linux ARM.
"av ; sys_platform == 'linux' and (platform_machine == 'aarch64' or platform_machine == 'arm64' or platform_machine == 'armv7l')",
@@ -87,15 +87,21 @@ url = "https://pypi.org/simple"
default = true
[[tool.uv.index]]
name = "torch-cu129"
url = "https://download.pytorch.org/whl/cu129"
name = "torch-cu130"
url = "https://download.pytorch.org/whl/cu130"
explicit = true
# To be removed after pypi sglang-kernel uses cu130
[[tool.uv.index]]
name = "sglang-kernel-cu130"
url = "https://docs.sglang.ai/whl/cu130/"
explicit = true
[tool.uv.sources]
torch = [
{ index = "pypi", marker = "platform_machine == 'x86_64'"},
{ index = "torch-cu129", marker = "platform_machine == 'aarch64'"},
]
torch = { index = "torch-cu130" }
torchvision = { index = "torch-cu130" }
torchaudio = { index = "torch-cu130" }
sglang-kernel = { index = "sglang-kernel-cu130" }
[project.optional-dependencies]
checkpoint-engine = ["checkpoint-engine==0.1.2"]
@@ -107,6 +113,7 @@ diffusion = [
"imageio==2.36.0",
"imageio-ffmpeg==0.5.1",
"moviepy>=2.0.0",
"nvidia-modelopt",
"opencv-python-headless==4.10.0.84",
"remote-pdb==2.1.0",
"st_attn==0.0.7 ; platform_machine != 'aarch64' and platform_machine != 'arm64'",