diff --git a/tools/server/server.cpp b/tools/server/server.cpp index 5293a98f034f1..60326e8e50efe 100644 --- a/tools/server/server.cpp +++ b/tools/server/server.cpp @@ -4226,7 +4226,7 @@ struct server_context { metrics.on_prompt_eval(slot); } - slot.t_token_generation = (t_current - slot.t_start_generation) / 1e3; + slot.t_token_generation = std::max(1, t_current - slot.t_start_generation) / 1e3; completion_token_output result; result.tok = id;