From 712e4d94502189362e012db7eaa09ba457c189d4 Mon Sep 17 00:00:00 2001 From: Joe Todd Date: Fri, 28 Jun 2024 13:29:00 +0100 Subject: [PATCH] Generate full token count during warm up Signed-off-by: Joe Todd --- examples/llama-bench/llama-bench.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/examples/llama-bench/llama-bench.cpp b/examples/llama-bench/llama-bench.cpp index d641a9f12..0f763c86c 100644 --- 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++) {