Generate full token count during warm up

Signed-off-by: Joe Todd <joe.todd@codeplay.com>
2024-06-28 13:29:00 +01:00 · 2024-06-28 13:29:00 +01:00 · 712e4d9450
parent 38373cfbab
commit 712e4d9450
1 changed files with 1 additions and 1 deletions
--- a/examples/llama-bench/llama-bench.cpp
+++ b/examples/llama-bench/llama-bench.cpp
@ -1390,7 +1390,7 @@ int main(int argc, char ** argv) {
            test_prompt(ctx, t.n_prompt, 0, t.n_batch, t.n_threads);
        }
        if (t.n_gen > 0) {
-            test_gen(ctx, 1, 0, t.n_threads);
+            test_gen(ctx, t.n_gen, 0, t.n_threads);
        }

        for (int i = 0; i < params.reps; i++) {