Update exllamav3 to v0.0.30, pin xformers

This commit is contained in:
oobabooga 2026-04-19 11:39:00 -07:00
parent eb596b8e0a
commit 1b3af51dc5

View file

@ -30,6 +30,7 @@ transformers==5.5.*
triton-windows==3.5.1.post24; platform_system == "Windows"
tqdm
wandb
xformers==0.0.33.post2
# Gradio
https://github.com/oobabooga/gradio/releases/download/4.37.2-custom.19/gradio-4.37.2+custom.19-py3-none-any.whl
@ -45,7 +46,7 @@ https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.119.0/llama
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.119.0/llama_cpp_binaries-0.119.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.119.0/ik_llama_cpp_binaries-0.119.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.119.0/ik_llama_cpp_binaries-0.119.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.28/exllamav3-0.0.28+cu128.torch2.9.0-cp313-cp313-win_amd64.whl; platform_system == "Windows" and python_version == "3.13"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.28/exllamav3-0.0.28+cu128.torch2.9.0-cp313-cp313-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.13"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.30/exllamav3-0.0.30+cu128.torch2.9.0-cp313-cp313-win_amd64.whl; platform_system == "Windows" and python_version == "3.13"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.30/exllamav3-0.0.30+cu128.torch2.9.0-cp313-cp313-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.13"
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"
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"