mikhail/llama.cpp-mtp-turboquant
1
0
Fork 0
Code Issues Pull Requests Actions 30 Packages Projects Releases Wiki Activity

common : delegate assistant continuation to underlying template handlers (#23089)

Browse Source 
* common : delegate assistant continuation to template handler

* server : implement echo parameter to exclude assistant prefill in the response

* server : fix tests for prefill

* server : use existing llama template

* cont : clean up
This commit is contained in:
Aldehir Rojas
2026-05-17 07:36:05 -04:00
committed by GitHub
parent a6d6183dbc
commit 39cf5d6191
10 changed files with 1112 additions and 191 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/server/server-task.h
+1 -5
View File
@@ -112,11 +112,7 @@ struct task_result_state {
const std::string oai_resp_message_id;
std::string oai_resp_fc_id; // function call ID for current args delta
task_result_state(const common_chat_parser_params & chat_parser_params)
: chat_parser_params(chat_parser_params)
, oai_resp_id("resp_" + random_string())
, oai_resp_reasoning_id("rs_" + random_string())
, oai_resp_message_id("msg_" + random_string()) {}
task_result_state(const common_chat_parser_params & chat_parser_params);
// parse partial tool calls and update the internal state
common_chat_msg update_chat_msg(