From fa7fe4eaaffab4c4ccac37504c77214fae08f0ad Mon Sep 17 00:00:00 2001
From: Nexesenex <124105151+Nexesenex@users.noreply.github.com>
Date: Sun, 28 Jun 2026 01:26:03 +0200
Subject: [PATCH] cuda-graph: GLM4_MOE - Don't load layer.nextn.embed_tokens

---
 src/llama-load-tensors.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/llama-load-tensors.cpp b/src/llama-load-tensors.cpp
index 9cea174da..44b3f28f6 100644
--- a/src/llama-load-tensors.cpp
+++ b/src/llama-load-tensors.cpp
@@ -2954,7 +2954,7 @@ bool create_tensors_helper::create_glm4_moe_tensors(const LLM_TN & tn) {
             layer.nextn.embed_tokens     = create_tensor(nextn_host_ctx,
                     tn(LLM_TENSOR_NEXTN_EMBED_TOKENS, "weight", final_layer),
                     { n_embd, n_vocab },
-                    flags | llama_model_loader::TENSOR_NOT_REQUIRED);
+                    flags | llama_model_loader::TENSOR_NOT_REQUIRED | llama_model_loader::TENSOR_SKIP);
             layer.nextn.enorm            = create_tensor(nextn_ctx,
                     tn(LLM_TENSOR_NEXTN_ENORM, "weight", final_layer),
                     { n_embd },