llama : remove KV cache defragmentation logic (#15473)

ggml-ci
This commit is contained in:
Georgi Gerganov authored and GitHub committed 2025-08-22 12:22:13 +03:00
1 parent ad5c975c2d
commit 9ebebef62f
16 files changed
+32 -440

No files matched your search

-1
View File
@@ -1152,7 +1152,6 @@ struct llama_context_params common_context_params_to_llama(const common_params &
cparams.yarn_orig_ctx = params.yarn_orig_ctx;
cparams.pooling_type = params.pooling_type;
cparams.attention_type = params.attention_type;
cparams.defrag_thold = params.defrag_thold;
cparams.cb_eval = params.cb_eval;
cparams.cb_eval_user_data = params.cb_eval_user_data;
cparams.offload_kqv = !params.no_kv_offload;