wmantly/llama.cpp-xrip
1
0 Fork 0
Code Issues Pull Requests Actions 27 Packages Projects Releases Wiki Activity

llama : optimize vocabulary loading (#30679)

Browse Source 
* llama : optimize vocabulary loading

* Update src/unicode.h

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* CR : added reserve in gguf_kv_to_str

* CR : trailing space fix

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
Evgen MykhailenkoandGeorgi Gerganov authored and GitHub committed 2026-10-21 12:03:27 +03:00
1 parent eb182d85c0
commit 824eaecf48
7 files changed
+75 -39
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

src/llama-model-loader.cpp
+1 -1
View File
@@ -807,8 +807,8 @@ llama_model_loader::llama_model_loader(
? format("%s[%s,%zu]", gguf_type_name(type), gguf_type_name(gguf_get_arr_type(metadata, i)), gguf_get_arr_n(metadata, i))
: gguf_type_name(type);
std::string value = gguf_kv_to_str(metadata, i);
const size_t MAX_VALUE_LEN = 40;
std::string value = gguf_kv_to_str(metadata, i, MAX_VALUE_LEN);
if (value.size() > MAX_VALUE_LEN) {
value = format("%s...", value.substr(0, MAX_VALUE_LEN - 3).c_str());
}