Distribute and run LLMs with a single file.
C++ cross-platformggufllama-cpplocal-ailocal-inferencelocal-llmopen-source-aisingle-file-executablespeech-to-text
diff --git a/ggml/src/ggml.c b/ggml/src/ggml.c
--- a/llama.cpp/ggml/src/ggml.c
+++ b/llama.cpp/ggml/src/ggml.c
@@ -556,11 +556,13 @@ void ggml_time_init(void) {
InitOnceExecuteOnce(&once, ggml_time_init_once, NULL, NULL);
}
int64_t ggml_time_ms(void) {
+ ggml_time_init(); // llamafile: GPU DLLs carry their own ggml.c; init lazily so timer_freq is never 0
LARGE_INTEGER t;
QueryPerformanceCounter(&t);
return (int64_t) (((double) (t.QuadPart - timer_start) / timer_freq) * 1000);
}
int64_t ggml_time_us(void) {
+ ggml_time_init(); // llamafile: see ggml_time_ms
LARGE_INTEGER t;
QueryPerformanceCounter(&t);
return (int64_t) (((double) (t.QuadPart - timer_start) / timer_freq) * 1000000);