mirror of
https://github.com/LostRuins/koboldcpp.git
synced 2026-09-09 14:29:15 +02:00
Foad Abo Dahood
5d806aa257
The spacing eviction in create_checkpoint() keeps the oldest checkpoint and erases every later one within checkpoint_min_step of it. For prompts shorter than checkpoint_min_step this drops the checkpoint at n_tokens - 4 that the next request resumes from, so hybrid/recurrent models re-prefill from the previous checkpoint instead. Apply the spacing rule only once the list is at n_ctx_checkpoints, and replace an existing checkpoint at the same n_tokens instead of appending a duplicate.