theroyallab/tabbyAPI
1
0
Fork 0
mirror of https://github.com/theroyallab/tabbyAPI.git synced 2026-05-11 08:20:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

Dependencies: Switch to mjun0812 flash-attn wheels

Browse Source
This commit is contained in:
turboderp
2026-05-03 00:01:29 +02:00
parent 6e97aa5fc1
commit 51b67595f4
1 changed files with 10 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
pyproject.toml
View File

@@ -88,17 +88,17 @@ cu12 = [
"exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.31/exllamav3-0.0.31+cu128.torch2.9.0-cp311-cp311-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.11'",
"exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.31/exllamav3-0.0.31+cu128.torch2.9.0-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'",
# Windows FA2 from https://github.com/kingbri1/flash-attention/releases
"flash_attn @ https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp313-cp313-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.13'",
"flash_attn @ https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp312-cp312-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.12'",
"flash_attn @ https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp311-cp311-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.11'",
"flash_attn @ https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp310-cp310-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.10'",
# Windows FA2 from https://github.com/mjun0812/flash-attention-prebuild-wheels/releases
"flash_attn @ https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.7.13/flash_attn-2.8.3+cu128torch2.9-cp313-cp313-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.13'",
"flash_attn @ https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.7.13/flash_attn-2.8.3+cu128torch2.9-cp312-cp312-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.12'",
"flash_attn @ https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.7.13/flash_attn-2.8.3+cu128torch2.9-cp311-cp311-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.11'",
"flash_attn @ https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.7.13/flash_attn-2.8.3+cu128torch2.9-cp310-cp310-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.10'",
# Linux FA2 from https://github.com/kingbri1/flash-attention/releases
"flash_attn @ https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp313-cp313-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.13'",
"flash_attn @ https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp312-cp312-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.12'",
"flash_attn @ https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp311-cp311-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.11'",
"flash_attn @ https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'",
# Linux FA2 from https://github.com/mjun0812/flash-attention-prebuild-wheels/releases
"flash_attn @ https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.9.0/flash_attn-2.8.3+cu128torch2.9-cp313-cp313-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.13'",
"flash_attn @ https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.9.0/flash_attn-2.8.3+cu128torch2.9-cp312-cp312-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.12'",
"flash_attn @ https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.9.0/flash_attn-2.8.3+cu128torch2.9-cp311-cp311-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.11'",
"flash_attn @ https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.9.0/flash_attn-2.8.3+cu128torch2.9-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'",
]
amd = [
# Torch triton for ROCm