Merge branch 'main' of https://github.com/abetlen/llama-cpp-python into batch-processing

abetlen · abetlen · commit 7a1c2b5d2ee0 · 2024-01-12T02:18:21.000-05:00
diff --git a/llama_cpp/llama_cpp.py b/llama_cpp/llama_cpp.py
@@ -180,6 +180,8 @@ def _load_shared_library(lib_base_name: str):
 #     LLAMA_FTYPE_MOSTLY_Q5_K_M        = 17, // except 1d tensors
 #     LLAMA_FTYPE_MOSTLY_Q6_K          = 18, // except 1d tensors
 #     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_GUESSED = 1024, // not specified in the model file
 # };
@@ -200,6 +202,9 @@ def _load_shared_library(lib_base_name: str):
 LLAMA_FTYPE_MOSTLY_Q5_K_S = 16
 LLAMA_FTYPE_MOSTLY_Q5_K_M = 17
 LLAMA_FTYPE_MOSTLY_Q6_K = 18
+LLAMA_FTYPE_MOSTLY_IQ2_XXS = 19
+LLAMA_FTYPE_MOSTLY_IQ2_XS = 20
+LLAMA_FTYPE_MOSTLY_Q2_K_S = 21
 LLAMA_FTYPE_GUESSED = 1024
 
 # enum llama_rope_scaling_type {
diff --git a/vendor/llama.cpp b/vendor/llama.cpp
@@ -1 +1 @@
-Subproject commit 6efb8eb30e7025b168f3fda3ff83b9b386428ad6
+Subproject commit 1d118386fea031f01550f8cd47a5c86296e5333f
