sdgoij/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-05-11 19:44:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

spec : refactor

Browse Source
This commit is contained in:
Georgi Gerganov
2026-01-25 17:15:46 +02:00
parent af382c384a
commit 924517dd38
10 changed files with 217 additions and 240 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tools/server/server-task.cpp
View File

@@ -238,6 +238,8 @@ task_params server_task::params_from_json_cmpl(
params.speculative.n_min = json_value(data, "speculative.n_min", defaults.speculative.n_min);
params.speculative.n_max = json_value(data, "speculative.n_max", defaults.speculative.n_max);
params.speculative.p_min = json_value(data, "speculative.p_min", defaults.speculative.p_min);
// TODO: is this needed? remove?
//params.speculative.self_mode = json_value(data, "speculative.self_mode", defaults.speculative.self_mode);
//params.speculative.self_cfg = json_value(data, "speculative.self_cfg", defaults.speculative.self_cfg);
// Set params.speculative.configs. Use json-array "speculative.configs" if provided in data, otherwise use {}