From f84bd67c804d8c5923d21cb7aa21dbb50fba9fe6 Mon Sep 17 00:00:00 2001
From: JJJYmmm <1650675829@qq.com>
Date: Sun, 26 Oct 2025 20:00:33 +0800
Subject: [PATCH] bugfix: fix the arch check for qwen3vl-moe.

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

diff --git a/src/llama-model.cpp b/src/llama-model.cpp
index 8450fae9b2..d97c97737b 100644
--- a/src/llama-model.cpp
+++ b/src/llama-model.cpp
@@ -3350,7 +3350,7 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
                 {
                     // for deepstack features, we consider the embd to be [main_embd, deepstack_embd_1, deepstack_embd_2, ...]
                     int64_t n_embd = hparams.n_embd;
-                    if (arch == LLM_ARCH_QWEN3VL) { 
+                    if (arch == LLM_ARCH_QWEN3VLMOE) { 
                         n_embd = hparams.n_embd / (hparams.n_deepstack_layers + 1);
                     }
                     tok_embd = create_tensor(tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, 0);