wylab/llama.cpp
0
0
Fork 1
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-09-15 18:13:29 +02:00
Code Issues Packages Projects Releases Wiki Activity

default draft model devices to global devices value

Browse Source
This commit is contained in:
Ruben Ortlam
2026-09-06 07:58:40 +02:00
parent 477ecf18ef
commit cf1eb320aa
1 changed files with 4 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
common/speculative.cpp
+4 -1
View File
@@ -2467,7 +2467,10 @@ common_params common_base_params_to_speculative(const common_params & params) {
result.pooling_type = LLAMA_POOLING_TYPE_UNSPECIFIED;
if (has_draft) {
result.devices = params_spec.devices;
// default to global devices value
if (!params_spec.devices.empty()) {
result.devices = params_spec.devices;
}
result.model = params_spec.mparams;
result.n_gpu_layers = params_spec.n_gpu_layers;
result.tensor_buft_overrides = params_spec.tensor_buft_overrides;