From 5c94894a1a5930db9b95dd691ecce5994d41f755 Mon Sep 17 00:00:00 2001 From: kingbri Date: Sat, 30 Mar 2024 16:58:24 -0400 Subject: [PATCH] Dependencies: Update Flash Attention v2.5.6 Signed-off-by: kingbri --- pyproject.toml | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/pyproject.toml b/pyproject.toml index 0ea59bb..ce95cdf 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -53,12 +53,12 @@ cu121 = [ "exllamav2 @ https://github.com/turboderp/exllamav2/releases/download/v0.0.16/exllamav2-0.0.16+cu121-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/bdashore3/flash-attention/releases - "flash_attn @ https://github.com/bdashore3/flash-attention/releases/download/v2.5.2/flash_attn-2.5.2+cu122torch2.2.0cxx11abiFALSE-cp311-cp311-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.11'", - "flash_attn @ https://github.com/bdashore3/flash-attention/releases/download/v2.5.2/flash_attn-2.5.2+cu122torch2.2.0cxx11abiFALSE-cp310-cp310-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.10'", + "flash_attn @ https://github.com/bdashore3/flash-attention/releases/download/v2.5.6/flash_attn-2.5.6+cu122torch2.2.2cxx11abiFALSE-cp311-cp311-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.11'", + "flash_attn @ https://github.com/bdashore3/flash-attention/releases/download/v2.5.6/flash_attn-2.5.6+cu122torch2.2.2cxx11abiFALSE-cp310-cp310-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.10'", # Linux FA2 from https://github.com/Dao-AILab/flash-attention/releases - "flash_attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.2/flash_attn-2.5.2+cu122torch2.2cxx11abiFALSE-cp311-cp311-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.11'", - "flash_attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.2/flash_attn-2.5.2+cu122torch2.2cxx11abiFALSE-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'", + "flash_attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.6/flash_attn-2.5.6+cu122torch2.2cxx11abiFALSE-cp311-cp311-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.11'", + "flash_attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.6/flash_attn-2.5.6+cu122torch2.2cxx11abiFALSE-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'", ] cu118 = [ # Torch @@ -74,8 +74,8 @@ cu118 = [ "exllamav2 @ https://github.com/turboderp/exllamav2/releases/download/v0.0.16/exllamav2-0.0.16+cu118-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/Dao-AILab/flash-attention/releases - "flash_attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.2/flash_attn-2.5.2+cu118torch2.2cxx11abiFALSE-cp311-cp311-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.11'", - "flash_attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.2/flash_attn-2.5.2+cu118torch2.2cxx11abiFALSE-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'", + "flash_attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.6/flash_attn-2.5.6+cu118torch2.2cxx11abiFALSE-cp311-cp311-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.11'", + "flash_attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.6/flash_attn-2.5.6+cu118torch2.2cxx11abiFALSE-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'", ] amd = [ # Torch triton for ROCm