Dependencies: Update flash attention 2

v2.5.8

Signed-off-by: kingbri <[email protected]>

pyproject.toml

@@ -54,12 +54,12 @@ cu121 = [
     "exllamav2 @ https://github.com/turboderp/exllamav2/releases/download/v0.0.20/exllamav2-0.0.20+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/v2.5.6/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'",
+    "flash_attn @ https://github.com/bdashore3/flash-attention/releases/download/v2.5.8/flash_attn-2.5.8+cu122torch2.3.0cxx11abiFALSE-cp311-cp311-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.11'",
+    "flash_attn @ https://github.com/bdashore3/flash-attention/releases/download/v2.5.8/flash_attn-2.5.8+cu122torch2.3.0cxx11abiFALSE-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/v2.5.6/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'",
+    "flash_attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.8/flash_attn-2.5.8+cu122torch2.3cxx11abiFALSE-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/v2.5.8/flash_attn-2.5.8+cu122torch2.3cxx11abiFALSE-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'",
 ]
 cu118 = [
     # Torch
@@ -75,8 +75,8 @@ cu118 = [
     "exllamav2 @ https://github.com/turboderp/exllamav2/releases/download/v0.0.20/exllamav2-0.0.20+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/v2.5.6/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'",
+    "flash_attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.8/flash_attn-2.5.8+cu118torch2.3cxx11abiFALSE-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/v2.5.8/flash_attn-2.5.8+cu118torch2.3cxx11abiFALSE-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'",
 ]
 amd = [
     # Torch triton for ROCm