From: Georgi Gerganov <redacted>
Date: Fri, 5 Jun 2026 14:11:42 +0000 (+0300)
Subject: model : fix llama_model::n_gpu_layers() (#24188)
X-Git-Tag: upstream/0.0.10438~909
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=96fbe0039337a999613a983d66e2bfcc4bb554d7;p=pkg%2Fggml%2Fsources%2Fllama.cpp

model : fix llama_model::n_gpu_layers() (#24188)
---

diff --git a/src/llama-model.cpp b/src/llama-model.cpp
index c98cb27e4..1f442d8a3 100644
--- a/src/llama-model.cpp
+++ b/src/llama-model.cpp
@@ -1636,7 +1636,8 @@ const float * llama_model::tensor_split() const {
 }
 
 uint32_t llama_model::n_gpu_layers() const {
-    return params.n_gpu_layers >= 0 ? params.n_gpu_layers : hparams.n_layer() + 1;
+    // note: plus 1 for the "output" layer
+    return params.n_gpu_layers >= 0 ? params.n_gpu_layers : hparams.n_layer_all + 1;
 }
 
 llama_split_mode llama_model::split_mode() const {