From 86221cf6dace86f47d896a38e0de652db4aa81a8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Johannes=20G=C3=A4=C3=9Fler?= <johannesg@5d6.de>
Date: Wed, 1 Apr 2026 21:28:19 +0200
Subject: [PATCH] CUDA: fix FA kernel selection logic (#21271)

---
 ggml/src/ggml-cuda/fattn.cu | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/ggml/src/ggml-cuda/fattn.cu b/ggml/src/ggml-cuda/fattn.cu
index a21c536104..addf93205e 100644
--- a/ggml/src/ggml-cuda/fattn.cu
+++ b/ggml/src/ggml-cuda/fattn.cu
@@ -340,7 +340,14 @@ static best_fattn_kernel ggml_cuda_get_best_fattn_kernel(const int device, const
         case 128:
         case 112:
         case 256:
+            if (V->ne[0] != K->ne[0]) {
+                return BEST_FATTN_KERNEL_NONE;
+            }
+            break;
         case 512:
+            if (V->ne[0] != K->ne[0]) {
+                return BEST_FATTN_KERNEL_NONE;
+            }
             if (!gqa_opt_applies) {
                 return BEST_FATTN_KERNEL_NONE;
             }