Nicholai/jan
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: prebundle cudart and cublas

Browse Source
This commit is contained in:
Louis 2024-10-23 18:30:25 +07:00
parent dc87f37a9b
commit 5f075c8554
No known key found for this signature in database
GPG Key ID: 44FA9F4D33C37DE2
3 changed files with 13 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
extensions/inference-cortex-extension/download.bat
View File

@ -18,8 +18,8 @@ call .\node_modules\.bin\download %DOWNLOAD_URL%-avx.tar.gz -e --strip 1 -o %BIN
call .\node_modules\.bin\download %DOWNLOAD_URL%-avx2.tar.gz -e --strip 1 -o %BIN_PATH%/avx2/engines/cortex.llamacpp call .\node_modules\.bin\download %DOWNLOAD_URL%-avx2.tar.gz -e --strip 1 -o %BIN_PATH%/avx2/engines/cortex.llamacpp
call .\node_modules\.bin\download %DOWNLOAD_URL%-avx512.tar.gz -e --strip 1 -o %BIN_PATH%/avx512/engines/cortex.llamacpp call .\node_modules\.bin\download %DOWNLOAD_URL%-avx512.tar.gz -e --strip 1 -o %BIN_PATH%/avx512/engines/cortex.llamacpp
call .\node_modules\.bin\download %DOWNLOAD_URL%-vulkan.tar.gz -e --strip 1 -o %BIN_PATH%/vulkan/engines/cortex.llamacpp call .\node_modules\.bin\download %DOWNLOAD_URL%-vulkan.tar.gz -e --strip 1 -o %BIN_PATH%/vulkan/engines/cortex.llamacpp
call .\node_modules\.bin\download %CUDA_DOWNLOAD_URL%-cuda-12-0-windows-amd64.tar.gz -e --strip 1 -o %BIN_PATH% call .\node_modules\.bin\download %CUDA_DOWNLOAD_URL%/cuda-12-0-windows-amd64.tar.gz -e --strip 1 -o %BIN_PATH%
call .\node_modules\.bin\download %CUDA_DOWNLOAD_URL%-cuda-11-7-windows-amd64.tar.gz -e --strip 1 -o %BIN_PATH% call .\node_modules\.bin\download %CUDA_DOWNLOAD_URL%/cuda-11-7-windows-amd64.tar.gz -e --strip 1 -o %BIN_PATH%
@REM Loop through each folder and move DLLs (excluding engine.dll) @REM Loop through each folder and move DLLs (excluding engine.dll)

4
extensions/inference-cortex-extension/download.sh
View File

@ -23,8 +23,8 @@ if [ "$OS_TYPE" == "Linux" ]; then
download "${ENGINE_DOWNLOAD_URL}-linux-amd64-noavx-cuda-12-0.tar.gz" -e --strip 1 -o "./bin/noavx-cuda-12-0/engines/cortex.llamacpp" 1 download "${ENGINE_DOWNLOAD_URL}-linux-amd64-noavx-cuda-12-0.tar.gz" -e --strip 1 -o "./bin/noavx-cuda-12-0/engines/cortex.llamacpp" 1
download "${ENGINE_DOWNLOAD_URL}-linux-amd64-noavx-cuda-11-7.tar.gz" -e --strip 1 -o "./bin/noavx-cuda-11-7/engines/cortex.llamacpp" 1 download "${ENGINE_DOWNLOAD_URL}-linux-amd64-noavx-cuda-11-7.tar.gz" -e --strip 1 -o "./bin/noavx-cuda-11-7/engines/cortex.llamacpp" 1
download "${ENGINE_DOWNLOAD_URL}-linux-amd64-vulkan.tar.gz" -e --strip 1 -o "./bin/vulkan/engines/cortex.llamacpp" 1 download "${ENGINE_DOWNLOAD_URL}-linux-amd64-vulkan.tar.gz" -e --strip 1 -o "./bin/vulkan/engines/cortex.llamacpp" 1
download "${CUDA_DOWNLOAD_URL}-cuda-12-0-linux-amd64.tar.gz" -e --strip 1 -o "./bin" 1 download "${CUDA_DOWNLOAD_URL}/cuda-12-0-linux-amd64.tar.gz" -e --strip 1 -o "./bin" 1
download "${CUDA_DOWNLOAD_URL}-cuda-11-7-linux-amd64.tar.gz" -e --strip 1 -o "./bin" 1 download "${CUDA_DOWNLOAD_URL}/cuda-11-7-linux-amd64.tar.gz" -e --strip 1 -o "./bin" 1
elif [ "$OS_TYPE" == "Darwin" ]; then elif [ "$OS_TYPE" == "Darwin" ]; then
# macOS downloads # macOS downloads

18
extensions/inference-cortex-extension/src/node/execute.ts
View File

@ -91,15 +91,15 @@ export const executableCortexFile = (
: process.platform === 'darwin' : process.platform === 'darwin'
? os() ? os()
: [ : [
gpuRunMode(gpuSetting) !== 'cuda' || gpuRunMode(gpuSetting) !== 'cuda' ||
cpuInstruction === 'avx2' cpuInstruction === 'avx2' || cpuInstruction === 'avx512'
? cpuInstruction ? cpuInstruction
: 'noavx', : 'noavx',
gpuRunMode(gpuSetting), gpuRunMode(gpuSetting),
cudaVersion(gpuSetting), cudaVersion(gpuSetting),
] ]
.filter((e) => !!e) .filter((e) => !!e)
.join('-') .join('-')
let cudaVisibleDevices = gpuSetting?.gpus_in_use.join(',') ?? '' let cudaVisibleDevices = gpuSetting?.gpus_in_use.join(',') ?? ''
let vkVisibleDevices = gpuSetting?.gpus_in_use.join(',') ?? '' let vkVisibleDevices = gpuSetting?.gpus_in_use.join(',') ?? ''
let binaryName = `cortex-server${extension()}` let binaryName = `cortex-server${extension()}`