diff options
author | Georgi Gerganov <ggerganov@gmail.com> | 2023-06-13 20:20:07 +0300 |
---|---|---|
committer | GitHub <noreply@github.com> | 2023-06-13 20:20:07 +0300 |
commit | 2347e45e7bdb09c9a7d74b2c0bc86c2b65f0c343 (patch) | |
tree | f3fb417af8c27a472e82a171d9dfb47ff8e2034e /examples/main | |
parent | 74d4cfa3438cb58bd177eed30014e6588694aaa8 (diff) |
llama : do a warm-up eval at start for better timings (#1824)
Diffstat (limited to 'examples/main')
-rw-r--r-- | examples/main/main.cpp | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/examples/main/main.cpp b/examples/main/main.cpp index 66d5631..efa913e 100644 --- a/examples/main/main.cpp +++ b/examples/main/main.cpp @@ -331,6 +331,13 @@ int main(int argc, char ** argv) { std::vector<llama_token> embd; + // do one empty run to warm up the model + { + const std::vector<llama_token> tmp = { llama_token_bos(), }; + llama_eval(ctx, tmp.data(), tmp.size(), 0, params.n_threads); + llama_reset_timings(ctx); + } + while ((n_remain != 0 && !is_antiprompt) || params.interactive) { // predict if (embd.size() > 0) { |