From b39a7bf1b038e8d1c3d58a313228dfc4f3993f5c Mon Sep 17 00:00:00 2001
From: ravel7524 <58877666+ravel7524@users.noreply.github.com>
Date: Wed, 20 May 2026 03:52:21 +0200
Subject: [PATCH] ggml-cuda: tune RDNA3 Q6_K MMVQ nwarps (#23349)

---
 ggml/src/ggml-cuda/mmvq.cu | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/ggml/src/ggml-cuda/mmvq.cu b/ggml/src/ggml-cuda/mmvq.cu
index da48f313a38..73a0991e206 100644
--- a/ggml/src/ggml-cuda/mmvq.cu
+++ b/ggml/src/ggml-cuda/mmvq.cu
@@ -359,7 +359,9 @@ static constexpr __host__ __device__ int calc_nwarps(ggml_type type, int ncols_d
                 case GGML_TYPE_Q5_1:
                 case GGML_TYPE_Q8_0:
                 case GGML_TYPE_Q4_K:
+                    return 8;
                 case GGML_TYPE_Q6_K:
+                    return 2;
                 case GGML_TYPE_IQ4_NL:
                     return 8;
                 default: