From: Aman Gupta <redacted>
Date: Wed, 25 Mar 2026 13:17:27 +0000 (+0800)
Subject: llama-bench: print `-n-cpu-moe` when offloaded layers > 1 (#20984)
X-Git-Tag: upstream/0.0.8611~89
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=9c600bcd4b3b21f70c9d95cf8a938e43192eb492;p=pkg%2Fggml%2Fsources%2Fllama.cpp

llama-bench: print `-n-cpu-moe` when offloaded layers > 1 (#20984)
---

diff --git a/tools/llama-bench/llama-bench.cpp b/tools/llama-bench/llama-bench.cpp
index 25beb369e..0a23f6985 100644
--- a/tools/llama-bench/llama-bench.cpp
+++ b/tools/llama-bench/llama-bench.cpp
@@ -1807,7 +1807,7 @@ struct markdown_printer : public printer {
         if (!is_cpu_backend) {
             fields.emplace_back("n_gpu_layers");
         }
-        if (params.n_cpu_moe.size() > 1) {
+        if (params.n_cpu_moe.size() > 1 || params.n_cpu_moe != cmd_params_defaults.n_cpu_moe) {
             fields.emplace_back("n_cpu_moe");
         }
         if (params.n_threads.size() > 1 || params.n_threads != cmd_params_defaults.n_threads || is_cpu_backend) {