diff --git a/llama.cpp b/llama.cpp
index 86ded5ba8..440103ed3 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -4601,18 +4601,7 @@ static bool llm_load_tensors(
                     }
 
                     // output
-                    {
-                        model.output_norm = ml.create_tensor(ctx_output,       tn(LLM_TENSOR_OUTPUT_NORM, "weight"), {n_embd});
-                        if (model.arch != LLM_ARCH_MINICPM){
-                            model.output = ml.create_tensor(ctx_output_split, tn(LLM_TENSOR_OUTPUT, "weight"), {n_embd, n_vocab}, false);
-                            // if output is NULL, init from the input tok embed
-                            if (model.output == NULL) {
-                                model.output = ml.create_tensor(ctx_output, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab});
-                                ml.n_created--; // artificial tensor
-                                ml.size_data += ggml_nbytes(model.output);
-                            }
-                        }
-                    }
+                    model.output_norm = ml.create_tensor(ctx_output,       tn(LLM_TENSOR_OUTPUT_NORM, "weight"), {n_embd});
 
                     for (int i = 0; i < n_layer; ++i) {
                         ggml_context * ctx_layer = ctx_for_layer(i);