Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-08-29 08:30:51 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'upstream' into concedo_experimental

Browse Source 
# Conflicts:
#	.github/workflows/docker.yml
#	tests/test-chat.cpp
#	tools/mtmd/clip-impl.h
#	tools/server/README.md
This commit is contained in:
Concedo
2026-06-10 00:36:26 +08:00
parent fe13a2989d f71af352a5
commit cd49b19016
30 changed files with 893 additions and 483 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/completion/completion.cpp
+4 -9
View File
@@ -34,12 +34,8 @@
#endif
static llama_context ** g_ctx;
static llama_model ** g_model;
static common_sampler ** g_smpl;
static common_params * g_params;
static std::vector<llama_token> * g_input_tokens;
static std::ostringstream * g_output_ss;
static std::vector<llama_token> * g_output_tokens;
static bool is_interacting = false;
static bool need_insert_eot = false;
@@ -137,7 +133,6 @@ int llama_completion(int argc, char ** argv) {
llama_context * ctx = nullptr;
common_sampler * smpl = nullptr;
g_model = &model;
g_ctx = &ctx;
g_smpl = &smpl;
@@ -550,9 +545,9 @@ int llama_completion(int argc, char ** argv) {
int n_consumed = 0;
int n_session_consumed = 0;
std::vector<int> input_tokens; g_input_tokens = &input_tokens;
std::vector<int> output_tokens; g_output_tokens = &output_tokens;
std::ostringstream output_ss; g_output_ss = &output_ss;
std::vector<int> input_tokens;
std::vector<int> output_tokens;
std::ostringstream output_ss;
std::ostringstream assistant_ss; // for storing current assistant message, used in conversation mode
// the first thing we will do is to output the prompt, so set color accordingly
@@ -990,7 +985,7 @@ int llama_completion(int argc, char ** argv) {
LOG("\n%s: saving final output to session file '%s'\n", __func__, path_session.c_str());
session_tokens.insert(session_tokens.end(), embd.begin(), embd.end());
llama_state_save_file(ctx, path_session.c_str(), session_tokens.data(), session_tokens.size());
LOG_INF("saved final session to %s, n_tokens = %ld\n", path_session.data(), session_tokens.size());
LOG_INF("saved final session to %s, n_tokens = %zu\n", path_session.data(), session_tokens.size());
}