Commit 53ecea05a for llama.cpp
commit 53ecea05a0728e133478b06b3d543b6fcfaa418e
Author: JamePeng <jame_peng@sina.com>
Date: Fri Nov 20 02:11:11 2026 +0800
mtmd : fix YuE2 projector loading (#32606)
- Allow cls.output.weight to be absent from the mmproj. It is an optional
Qwen3 reranking head that the YuE2 NAR decoder never uses.
- Without this fix, mtmd_init_from_file() fails with
"tensor cls.output.weight is missing from the mmproj", preventing
audio generation from starting.
diff --git a/tools/mtmd/mtmd.cpp b/tools/mtmd/mtmd.cpp
index 1101d267a..511e01f93 100644
--- a/tools/mtmd/mtmd.cpp
+++ b/tools/mtmd/mtmd.cpp
@@ -644,6 +644,7 @@ static void mtmd_owned_llama_set_tensor_data(ggml_tensor * t, void * user_data)
return name.size() >= n && name.compare(name.size() - n, n, suffix) == 0;
};
const bool optional = ends_with("scale") || ends_with(".bias") || name == "dense_2.weight" ||
+ name == "cls.output.weight" || // Qwen3 reranking head, unused by YuE2 NAR
name == "rope_freqs.weight" || // granite, voxcpm2 residual LM
name == "token_embd.weight" || name == "output.weight"; // inputs are embedding rows, outputs are hidden states
if (!optional || t->type != GGML_TYPE_F32) {