Public/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-09-19 17:24:57 +02:00
Code Issues Packages Projects Releases Wiki Activity

Add a numeric context parameter declaring the maximum outputs one sequence

Browse Source
This commit is contained in:
Gaurav Garg
2026-07-05 02:48:13 +05:30
parent 0eddfa0b4c
commit eab4e8444f
14 changed files with 120 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/lookup/lookup.cpp
+1
View File
@@ -30,6 +30,7 @@ int main(int argc, char ** argv){
const int n_draft = params.speculative.draft.n_max;
params.n_outputs_max = common_speculative_n_outputs_max(params.n_batch, params.n_parallel, n_draft);
params.n_sampling_outputs_per_seq_max = common_speculative_n_outputs_per_seq_max(params.n_batch, n_draft);
// init llama.cpp
llama_backend_init();
examples/speculative-simple/speculative-simple.cpp
+5
View File
@@ -32,6 +32,8 @@ int main(int argc, char ** argv) {
params.n_outputs_max = common_speculative_n_outputs_max(
params.n_batch, params.n_parallel, common_speculative_n_max(&params.speculative));
params.n_sampling_outputs_per_seq_max = common_speculative_n_outputs_per_seq_max(
params.n_batch, common_speculative_n_max(&params.speculative));
// init llama.cpp
llama_backend_init();
@@ -59,6 +61,9 @@ int main(int argc, char ** argv) {
auto params_dft = params;
params_dft.n_outputs_max = params.n_parallel;
params_dft.n_sampling_outputs_per_seq_max = 1;
params_dft.devices = params_spec.devices;
params_dft.model = params_spec.mparams;
params_dft.n_gpu_layers = params_spec.n_gpu_layers;
examples/speculative/speculative.cpp
+4
View File
@@ -60,6 +60,8 @@ int main(int argc, char ** argv) {
params.n_outputs_max = common_speculative_n_outputs_max(
params.n_batch, params.n_parallel, params.speculative.draft.n_max);
params.n_sampling_outputs_per_seq_max = common_speculative_n_outputs_per_seq_max(
params.n_batch, params.speculative.draft.n_max);
// probability threshold for splitting a draft branch (only for n_seq_dft > 1)
const float p_draft_split = params.speculative.draft.p_split;
@@ -87,6 +89,8 @@ int main(int argc, char ** argv) {
params.devices = params.speculative.draft.devices;
params.model = params.speculative.draft.mparams;
params.n_gpu_layers = params.speculative.draft.n_gpu_layers;
params.n_outputs_max = params.n_parallel;
params.n_sampling_outputs_per_seq_max = 1;
if (params.speculative.draft.cpuparams.n_threads > 0) {
params.cpuparams.n_threads = params.speculative.draft.cpuparams.n_threads;
}