From 7a63a560430641d55b685994c097ab7a588b6547 Mon Sep 17 00:00:00 2001 From: oobabooga <112222186+oobabooga@users.noreply.github.com> Date: Wed, 11 Mar 2026 12:53:19 -0700 Subject: [PATCH] Update llama.cpp --- requirements/full/requirements.txt | 4 ++-- requirements/full/requirements_amd.txt | 4 ++-- requirements/full/requirements_apple_intel.txt | 2 +- requirements/full/requirements_apple_silicon.txt | 2 +- requirements/full/requirements_cpu_only.txt | 4 ++-- requirements/portable/requirements.txt | 4 ++-- requirements/portable/requirements_amd.txt | 4 ++-- requirements/portable/requirements_apple_intel.txt | 2 +- requirements/portable/requirements_apple_silicon.txt | 2 +- requirements/portable/requirements_cpu_only.txt | 4 ++-- requirements/portable/requirements_cuda131.txt | 4 ++-- requirements/portable/requirements_vulkan.txt | 4 ++-- 12 files changed, 20 insertions(+), 20 deletions(-) diff --git a/requirements/full/requirements.txt b/requirements/full/requirements.txt index a3d4d1e6..057ff94e 100644 --- a/requirements/full/requirements.txt +++ b/requirements/full/requirements.txt @@ -40,8 +40,8 @@ sse-starlette==1.6.5 tiktoken # CUDA wheels -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows" -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/turboderp-org/exllamav3/releases/download/v0.0.24/exllamav3-0.0.24+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/v0.0.24/exllamav3-0.0.24+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" diff --git a/requirements/full/requirements_amd.txt b/requirements/full/requirements_amd.txt index 3211f251..19c45361 100644 --- a/requirements/full/requirements_amd.txt +++ b/requirements/full/requirements_amd.txt @@ -37,5 +37,5 @@ sse-starlette==1.6.5 tiktoken # AMD wheels -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+rocm6.4-py3-none-win_amd64.whl; platform_system == "Windows" -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+rocm6.4-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+rocm6.4-py3-none-win_amd64.whl; platform_system == "Windows" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+rocm6.4-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" diff --git a/requirements/full/requirements_apple_intel.txt b/requirements/full/requirements_apple_intel.txt index 8d452114..4f8ea026 100644 --- a/requirements/full/requirements_apple_intel.txt +++ b/requirements/full/requirements_apple_intel.txt @@ -37,4 +37,4 @@ sse-starlette==1.6.5 tiktoken # Mac wheels -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin" diff --git a/requirements/full/requirements_apple_silicon.txt b/requirements/full/requirements_apple_silicon.txt index 525ceed5..9b5bf1ab 100644 --- a/requirements/full/requirements_apple_silicon.txt +++ b/requirements/full/requirements_apple_silicon.txt @@ -37,4 +37,4 @@ sse-starlette==1.6.5 tiktoken # Mac wheels -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin" diff --git a/requirements/full/requirements_cpu_only.txt b/requirements/full/requirements_cpu_only.txt index 86b65a97..7e49bbff 100644 --- a/requirements/full/requirements_cpu_only.txt +++ b/requirements/full/requirements_cpu_only.txt @@ -37,5 +37,5 @@ sse-starlette==1.6.5 tiktoken # llama.cpp (CPU only) -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows" diff --git a/requirements/portable/requirements.txt b/requirements/portable/requirements.txt index 61c9ef73..00c6d0d6 100644 --- a/requirements/portable/requirements.txt +++ b/requirements/portable/requirements.txt @@ -23,5 +23,5 @@ sse-starlette==1.6.5 tiktoken # CUDA wheels -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows" -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" diff --git a/requirements/portable/requirements_amd.txt b/requirements/portable/requirements_amd.txt index 3d0785a3..2d8dad0a 100644 --- a/requirements/portable/requirements_amd.txt +++ b/requirements/portable/requirements_amd.txt @@ -23,5 +23,5 @@ sse-starlette==1.6.5 tiktoken # AMD wheels -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+rocm6.4-py3-none-win_amd64.whl; platform_system == "Windows" -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+rocm6.4-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+rocm6.4-py3-none-win_amd64.whl; platform_system == "Windows" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+rocm6.4-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" diff --git a/requirements/portable/requirements_apple_intel.txt b/requirements/portable/requirements_apple_intel.txt index 6805e209..299f50c1 100644 --- a/requirements/portable/requirements_apple_intel.txt +++ b/requirements/portable/requirements_apple_intel.txt @@ -23,4 +23,4 @@ sse-starlette==1.6.5 tiktoken # Mac wheels -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin" diff --git a/requirements/portable/requirements_apple_silicon.txt b/requirements/portable/requirements_apple_silicon.txt index 5a8ed87b..dbc9247c 100644 --- a/requirements/portable/requirements_apple_silicon.txt +++ b/requirements/portable/requirements_apple_silicon.txt @@ -23,4 +23,4 @@ sse-starlette==1.6.5 tiktoken # Mac wheels -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin" diff --git a/requirements/portable/requirements_cpu_only.txt b/requirements/portable/requirements_cpu_only.txt index fafa23cf..66a99ba6 100644 --- a/requirements/portable/requirements_cpu_only.txt +++ b/requirements/portable/requirements_cpu_only.txt @@ -23,5 +23,5 @@ sse-starlette==1.6.5 tiktoken # llama.cpp (CPU only) -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows" diff --git a/requirements/portable/requirements_cuda131.txt b/requirements/portable/requirements_cuda131.txt index 3ef59f97..4338b20c 100644 --- a/requirements/portable/requirements_cuda131.txt +++ b/requirements/portable/requirements_cuda131.txt @@ -23,5 +23,5 @@ sse-starlette==1.6.5 tiktoken # CUDA wheels -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+cu131-py3-none-win_amd64.whl; platform_system == "Windows" -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+cu131-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu131-py3-none-win_amd64.whl; platform_system == "Windows" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu131-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" diff --git a/requirements/portable/requirements_vulkan.txt b/requirements/portable/requirements_vulkan.txt index 6039357d..7d262302 100644 --- a/requirements/portable/requirements_vulkan.txt +++ b/requirements/portable/requirements_vulkan.txt @@ -23,5 +23,5 @@ sse-starlette==1.6.5 tiktoken # Vulkan wheels -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows" -https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.87.0/llama_cpp_binaries-0.87.0+vulkan-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows" +https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+vulkan-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"