lhutyra
diff --git a/‎llama_cpp/llama_cpp.py
Copy file name to clipboardExpand all lines: llama_cpp/llama_cpp.py
+2Lines changed: 2 additions & 0 deletions b/‎llama_cpp/llama_cpp.py
Copy file name to clipboardExpand all lines: llama_cpp/llama_cpp.py
+2Lines changed: 2 additions & 0 deletions
diff --git a/‎vendor/llama.cpp
Copy file name to clipboard b/‎vendor/llama.cpp
Copy file name to clipboard
@@ -188,6 +188,7 @@ def _load_shared_library(lib_base_name: str):
 #     LLAMA_FTYPE_MOSTLY_IQ2_XXS       = 19, // except 1d tensors
 #     LLAMA_FTYPE_MOSTLY_IQ2_XS        = 20, // except 1d tensors
 #     LLAMA_FTYPE_MOSTLY_Q2_K_S        = 21, // except 1d tensors
+#     LLAMA_FTYPE_MOSTLY_Q3_K_XS       = 22, // except 1d tensors
 
 #     LLAMA_FTYPE_GUESSED = 1024, // not specified in the model file
 # };
@@ -211,6 +212,7 @@ def _load_shared_library(lib_base_name: str):
 LLAMA_FTYPE_MOSTLY_IQ2_XXS = 19
 LLAMA_FTYPE_MOSTLY_IQ2_XS = 20
 LLAMA_FTYPE_MOSTLY_Q2_K_S = 21
+LLAMA_FTYPE_MOSTLY_Q3_K_XS = 22
 LLAMA_FTYPE_GUESSED = 1024
 
 # enum llama_rope_scaling_type {