mirror of
https://github.com/ggerganov/llama.cpp
synced 2026-04-29 02:32:01 +02:00
download : prefer q8_0 when q4_k not available
This commit is contained in:
parent
0f1bb602dd
commit
fd6f79c7a4
@ -627,7 +627,7 @@ static hf_cache::hf_file find_best_model(const hf_cache::hf_files & files,
|
||||
if (!tag.empty()) {
|
||||
tags.push_back(tag);
|
||||
} else {
|
||||
tags = {"Q4_K_M", "Q4_0"};
|
||||
tags = {"Q4_K_M", "Q8_0"};
|
||||
}
|
||||
|
||||
for (const auto & t : tags) {
|
||||
|
||||
Loading…
Reference in New Issue
Block a user