We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent fa654d0 commit bf31654Copy full SHA for bf31654
llama.cpp
@@ -3647,6 +3647,7 @@ static bool llm_load_tensors(
3647
model.buft_layer[i] = llama_default_buffer_type_cpu(true);
3648
}
3649
3650
+#ifndef GGML_USE_KOMPUTE
3651
if (split_mode == LLAMA_SPLIT_LAYER) {
3652
// calculate the split points
3653
int device_count = llama_get_device_count();
@@ -3684,7 +3685,9 @@ static bool llm_load_tensors(
3684
3685
} else {
3686
model.buft_output = llama_default_buffer_type_cpu(true);
3687
- } else {
3688
+ } else
3689
+#endif
3690
+ {
3691
ggml_backend_buffer_type_t split_buft;
3692
if (split_mode == LLAMA_SPLIT_ROW) {
3693
split_buft = llama_default_buffer_type_split(main_gpu, tensor_split);
0 commit comments