From 7125772c0adff14176e201d0da1d5f4bca3c03db Mon Sep 17 00:00:00 2001
From: Daniel Han <23090290+danielhanchen@users.noreply.github.com>
Date: Wed, 18 Mar 2026 06:16:37 +0000
Subject: [PATCH] Remove extra CUDA cmake flags to align Windows with Linux
 build

Drop GGML_CUDA_FA_ALL_QUANTS, GGML_CUDA_F16, GGML_CUDA_GRAPHS,
GGML_CUDA_FORCE_CUBLAS, and GGML_CUDA_PEER_MAX_BATCH_SIZE flags.
The Linux build in setup.sh only sets GGML_CUDA=ON and lets llama.cpp
use its defaults for everything else. Keep Windows consistent.
---
 studio/setup.ps1 | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/studio/setup.ps1 b/studio/setup.ps1
index 87bc43b87f..667460ebd3 100644
--- a/studio/setup.ps1
+++ b/studio/setup.ps1
@@ -1146,11 +1146,6 @@ if (Test-Path $LlamaServerBin) {
             $CmakeArgs += "-DCUDAToolkit_ROOT=$CudaToolkitRoot"
             $CmakeArgs += "-DCUDA_TOOLKIT_ROOT_DIR=$CudaToolkitRoot"
             $CmakeArgs += "-DCMAKE_CUDA_COMPILER=$NvccPath"
-            $CmakeArgs += '-DGGML_CUDA_FA_ALL_QUANTS=ON'
-            $CmakeArgs += '-DGGML_CUDA_F16=OFF'
-            $CmakeArgs += '-DGGML_CUDA_GRAPHS=OFF'
-            $CmakeArgs += '-DGGML_CUDA_FORCE_CUBLAS=OFF'
-            $CmakeArgs += '-DGGML_CUDA_PEER_MAX_BATCH_SIZE=8192'
             if ($CudaArch) {
                 # Validate nvcc actually supports this architecture
                 if (Test-NvccArchSupport -NvccExe $NvccPath -Arch $CudaArch) {