vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

server : fix draft context not being released

Browse source
This commit is contained in:
slaren 2025-01-22 17:24:49 +01:00
parent 96f4053934
commit 0410e03ced
1 changed files with 3 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

3
examples/server/server.cpp
View file

@ -1772,6 +1772,9 @@ struct server_context {
// force F16 KV cache for the draft model for extra performance // force F16 KV cache for the draft model for extra performance
cparams_dft.type_k = GGML_TYPE_F16; cparams_dft.type_k = GGML_TYPE_F16;
cparams_dft.type_v = GGML_TYPE_F16; cparams_dft.type_v = GGML_TYPE_F16;
// the context is not needed - we will create one for each slot
llama_init_dft.context.reset();
} }
chat_templates = common_chat_templates_from_model(model, params_base.chat_template); chat_templates = common_chat_templates_from_model(model, params_base.chat_template);