Use the token cache to infer greater n_past and reuse results (#3073)

Signed-off-by: Jared Van Bortel <jared@nomic.ai>
2025-09-08 20:09:12 +00:00 · 2024-10-31 11:19:12 -04:00
parent 62cab695eb
commit f07e2e63df
15 changed files with 320 additions and 169 deletions
--- a/common/common.cmake
+++ b/common/common.cmake
@@ -12,6 +12,7 @@ function(gpt4all_add_warning_options target)
        -Wformat=2
        -Wmissing-include-dirs
        -Wstrict-overflow=2
+        -Wsuggest-override
        -Wvla
        # errors
        -Werror=format-security