furyhawk/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-07-21 10:15:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

server: MTP layer kv-cache should respect draft type ctk (#23646)

Browse Source
This commit is contained in:
Aman Gupta
2026-05-25 16:46:23 +08:00
committed by GitHub
parent 5fdf07e33b
commit 6c4cbdc70b
1 changed files with 4 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/server/server-context.cpp
+4
View File
@@ -822,6 +822,8 @@ private:
auto cparams_dft = common_context_params_to_llama(params_dft);
if (spec_mtp) {
cparams_dft.ctx_type = LLAMA_CONTEXT_TYPE_MTP;
cparams_dft.type_k = params_base.speculative.draft.cache_type_k;
cparams_dft.type_v = params_base.speculative.draft.cache_type_v;
}
cparams_dft.n_rs_seq = 0;
@@ -940,6 +942,8 @@ private:
auto cparams_mtp = common_context_params_to_llama(params_base);
cparams_mtp.ctx_type = LLAMA_CONTEXT_TYPE_MTP;
cparams_mtp.type_k = params_base.speculative.draft.cache_type_k;
cparams_mtp.type_v = params_base.speculative.draft.cache_type_v;
cparams_mtp.n_rs_seq = 0;
ctx_dft.reset(llama_init_from_model(model_tgt, cparams_mtp));