From 186626d58e939992fdb5594acd00f5ddc3affba5 Mon Sep 17 00:00:00 2001
From: Andrei Betlen <abetlen@gmail.com>
Date: Fri, 1 Sep 2023 14:26:13 -0400
Subject: [PATCH] Update llama.cpp

---
 llama_cpp/llama_cpp.py | 2 ++
 vendor/llama.cpp       | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/llama_cpp/llama_cpp.py b/llama_cpp/llama_cpp.py
index 205f7eb..bf81518 100644
--- a/llama_cpp/llama_cpp.py
+++ b/llama_cpp/llama_cpp.py
@@ -294,6 +294,7 @@ llama_log_callback = ctypes.CFUNCTYPE(None, c_int, c_char_p, c_void_p)
 #     enum llama_ftype ftype;      // quantize to this llama_ftype
 #     bool allow_requantize;       // allow quantizing non-f32/f16 tensors
 #     bool quantize_output_tensor; // quantize output.weight
+#     bool only_copy;              // only copy tensors - ftype, allow_requantize and quantize_output_tensor are ignored
 # } llama_model_quantize_params;
 class llama_model_quantize_params(Structure):
     _fields_ = [
@@ -301,6 +302,7 @@ class llama_model_quantize_params(Structure):
         ("ftype", c_int),
         ("allow_requantize", c_bool),
         ("quantize_output_tensor", c_bool),
+        ("only_copy", c_bool),
     ]
 
 
diff --git a/vendor/llama.cpp b/vendor/llama.cpp
index bcce96b..69fdbb9 160000
--- a/vendor/llama.cpp
+++ b/vendor/llama.cpp
@@ -1 +1 @@
-Subproject commit bcce96ba4dd95482824700c4ce2455fe8c49055a
+Subproject commit 69fdbb9abc8907dd2a9ffdd840cba92d678a660a