diff --git a/src/llama-context.cpp b/src/llama-context.cpp
index 0402044da6..97fe39b7b1 100644
--- a/src/llama-context.cpp
+++ b/src/llama-context.cpp
@@ -2301,6 +2301,7 @@ uint32_t llama_context::graph_max_nodes(uint32_t n_tokens) const {
         model.arch == LLM_ARCH_BAILINGMOE3 ||
         model.arch == LLM_ARCH_QWEN35 ||
         model.arch == LLM_ARCH_QWEN35MOE ||
+        model.arch == LLM_ARCH_QWEN4EXP ||
         model.arch == LLM_ARCH_DEEPSEEK4 ||
         (model.arch == LLM_ARCH_DFLASH && model.hparams.dsv4_hc_mult > 0) ||
         model.arch == LLM_ARCH_NANBEIGE ||