model : print tensor size during load

ggerganov · ggerganov · commit 18b0e5678b1b · 2025-04-02T11:43:00.000+03:00
diff --git a/src/llama-model-loader.cpp b/src/llama-model-loader.cpp
@@ -600,7 +600,9 @@ llama_model_loader::llama_model_loader(
 
             if (trace > 0) {
                 const uint16_t sid = w.idx;
-                LLAMA_LOG_INFO("%s: - tensor split %2d: %32s %-8s [ %s ]\n", __func__, sid, ggml_get_name(tensor), ggml_type_name(type), llama_format_tensor_shape(tensor).c_str());
+                LLAMA_LOG_INFO("%s: - tensor split %2d: %32s %-8s [ %s ] %8.2f MB\n", __func__,
+                        sid, ggml_get_name(tensor), ggml_type_name(type), llama_format_tensor_shape(tensor).c_str(),
+                        ggml_nbytes(tensor)/1024.0f/1024.0f);
             }
         }
 

Original file line number	Diff line number	Diff line change
`@@ -600,7 +600,9 @@ llama_model_loader::llama_model_loader(`
`600`	`600`
`601`	`601`	`if (trace > 0) {`
`602`	`602`	`const uint16_t sid = w.idx;`
`603`		`- LLAMA_LOG_INFO("%s: - tensor split %2d: %32s %-8s [ %s ]\n", __func__, sid, ggml_get_name(tensor), ggml_type_name(type), llama_format_tensor_shape(tensor).c_str());`
	`603`	`+ LLAMA_LOG_INFO("%s: - tensor split %2d: %32s %-8s [ %s ] %8.2f MB\n", __func__,`
	`604`	`+ sid, ggml_get_name(tensor), ggml_type_name(type), llama_format_tensor_shape(tensor).c_str(),`
	`605`	`+ ggml_nbytes(tensor)/1024.0f/1024.0f);`
`604`	`606`	`}`
`605`	`607`	`}`
`606`	`608`