From 01763e800dfede49f2d4225b8a2df3c8f7993afc Mon Sep 17 00:00:00 2001 From: Sascha Rogmann Date: Tue, 10 Feb 2026 23:51:27 +0100 Subject: [PATCH] server : log levels --- tools/server/server-context.cpp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/server/server-context.cpp b/tools/server/server-context.cpp index f3db868611..dd8706d613 100644 --- a/tools/server/server-context.cpp +++ b/tools/server/server-context.cpp @@ -2149,7 +2149,7 @@ private: slot.prompt.tokens.push_back(slot.sampled); if (slot.task->params.speculative.n_min > (int) draft.size()) { - SLT_INF(slot, "ignoring small draft: %d < %d\n", (int) draft.size(), slot.task->params.speculative.n_min); + SLT_DBG(slot, "ignoring small draft: %d < %d\n", (int) draft.size(), slot.task->params.speculative.n_min); // fallback to normal decoding slot.i_batch = slot.i_batch_dft[0]; slot.drafted.clear(); @@ -2682,10 +2682,10 @@ private: // do not checkpoint after mtmd chunks do_checkpoint = do_checkpoint && !has_mtmd; - SLT_INF(slot, "main/do_checkpoint = %s, pos_min = %d, pos_max = %d\n", do_checkpoint ? "yes" : "no", pos_min, pos_max); // no need to create checkpoints that are too close together do_checkpoint = do_checkpoint && (slot.prompt.checkpoints.empty() || slot.prompt.n_tokens() - n_tokens_cur > slot.prompt.checkpoints.back().n_tokens + 64); + SLT_DBG(slot, "main/do_checkpoint = %s, pos_min = %d, pos_max = %d\n", do_checkpoint ? "yes" : "no", pos_min, pos_max); // note: we create the checkpoint before calling llama_decode(), so the current batch is not // yet processed and therefore it is not part of the checkpoint.