From d417bc43dd29eab006a0da73afc7d610c9ebae7d Mon Sep 17 00:00:00 2001 From: Georgi Gerganov Date: Sun, 8 Mar 2026 22:16:46 +0200 Subject: [PATCH] server : do not create checkpoints right after mtmd chunks (#20232) --- tools/server/server-context.cpp | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/tools/server/server-context.cpp b/tools/server/server-context.cpp index 9dbd6d798a..3790308d0e 100644 --- a/tools/server/server-context.cpp +++ b/tools/server/server-context.cpp @@ -2438,6 +2438,8 @@ private: slot.n_prompt_tokens_cache = 0; } + bool do_checkpoint = params_base.n_ctx_checkpoints > 0; + // check if we should process the image if (slot.prompt.n_tokens() < slot.task->n_tokens() && input_tokens[slot.prompt.n_tokens()] == LLAMA_TOKEN_NULL) { // process the image @@ -2457,6 +2459,8 @@ private: const auto & chunk = input_tokens.find_chunk(slot.prompt.n_tokens()); slot.prompt.tokens.push_back(chunk.get()); // copy } + + do_checkpoint = false; // do not checkpoint right after an image chunk } // If using an alora, there may be uncached tokens that come @@ -2473,8 +2477,6 @@ private: alora_disabled_id = enabled_loras[0]; } - bool do_checkpoint = params_base.n_ctx_checkpoints > 0; - // make checkpoints only for completion tasks do_checkpoint = do_checkpoint && slot.task->type == SERVER_TASK_TYPE_COMPLETION;