From: Sigbjørn Skjæret <redacted>
Date: Thu, 25 Jun 2026 05:36:49 +0000 (+0200)
Subject: quant : fix quantizing moe with mtp (#24986)
X-Git-Tag: upstream/0.0.10438~649
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=b3ce5cedf4c007b78a45befe839fa3abada03c0b;p=pkg%2Fggml%2Fsources%2Fllama.cpp

quant : fix quantizing moe with mtp (#24986)
---

diff --git a/src/llama-quant.cpp b/src/llama-quant.cpp
index 89b7fe8d4..847e79f46 100644
--- a/src/llama-quant.cpp
+++ b/src/llama-quant.cpp
@@ -847,7 +847,7 @@ static void init_quantize_state_counters(quantize_state_impl & qs, std::vector<t
             qs.has_tied_embeddings = false;
         }
     }
-    qs.n_ffn_down = qs.n_ffn_gate = qs.n_ffn_up = (int)qs.model.hparams.n_layer();
+    qs.n_ffn_down = qs.n_ffn_gate = qs.n_ffn_up = (int)qs.model.hparams.n_layer_all;
 }
 
 //