diff --git a/ggml-cuda.cu b/ggml-cuda.cu
index 46ac296be..1df0c578e 100644
--- a/ggml-cuda.cu
+++ b/ggml-cuda.cu
@@ -4563,7 +4563,6 @@ struct cuda_buffer {
 
 static cuda_buffer g_cuda_buffer_pool[GGML_CUDA_MAX_DEVICES][MAX_CUDA_BUFFERS];
 static std::atomic_flag g_cuda_pool_lock = ATOMIC_FLAG_INIT;
-static bool g_mul_mat_q = false;
 
 static void * ggml_cuda_pool_malloc(size_t size, size_t * actual_size) {
     scoped_spin_lock lock(g_cuda_pool_lock);
diff --git a/koboldcpp.py b/koboldcpp.py
index c4ce973fc..ab3ce85ec 100755
--- a/koboldcpp.py
+++ b/koboldcpp.py
@@ -303,7 +303,7 @@ maxhordectx = 1024
 maxhordelen = 256
 modelbusy = threading.Lock()
 defaultport = 5001
-KcppVersion = "1.40"
+KcppVersion = "1.40.1"
 showdebug = True
 showsamplerwarning = True
 showmaxctxwarning = True