Distribute and run LLMs with a single file.
C++ cross-platformggufllama-cpplocal-ailocal-inferencelocal-llmopen-source-aisingle-file-executablespeech-to-text
Add --version flag
Justine Tunneyd4e23880ddd1
diff --git a/llama.cpp/common.h b/llama.cpp/common.hindex bebe5b82..e9e1093f 100644--- a/llama.cpp/common.h+++ b/llama.cpp/common.h@@ -3,8 +3,8 @@#pragma once#include "llama.h"-#include "sampling.h"+#include "llamafile/version.h"#define LOG_NO_FILE_LINE_FUNCTION#include "log.h"@@ -26,9 +26,8 @@#define die(msg) do { fputs("error: " msg "\n", stderr); exit(1); } while (0)#define die_fmt(fmt, ...) do { fprintf(stderr, "error: " fmt "\n", __VA_ARGS__); exit(1); } while (0)-#define print_build_info() do { \- fprintf(stderr, "%s: build = %d (%s)\n", __func__, LLAMA_BUILD_NUMBER, LLAMA_COMMIT); \- fprintf(stderr, "%s: built with %s for %s\n", __func__, LLAMA_COMPILER, LLAMA_BUILD_TARGET); \+#define print_build_info() do { \+ fprintf(stderr, "%s: llamafile version " LLAMAFILE_VERSION_STRING "\n", __func__, LLAMA_BUILD_NUMBER, LLAMA_COMMIT); \} while(0)// build infodiff --git a/llama.cpp/main/main.cc b/llama.cpp/main/main.ccindex 801891b1..f3760935 100644--- a/llama.cpp/main/main.cc+++ b/llama.cpp/main/main.cc@@ -4,6 +4,7 @@#include <cosmo.h>#include "llama.cpp/ggml-metal.h"#include "llama.cpp/ggml-cuda.h"+#include "llamafile/version.h"#include "tool/args/args.h"#include "llama.cpp/common.h"@@ -105,6 +106,11 @@ static int Eval(struct llama_context * ctx, struct llama_batch batch) {}int main(int argc, char ** argv) {+ if (argc == 2 && !strcmp(argv[1], "--version")) {+ printf("llamafile v" LLAMAFILE_VERSION_STRING " main\n");+ exit(0);+ }+llamafile_check_cpu();ShowCrashReports();LoadZipArgs(&argc, &argv);@@ -184,8 +190,7 @@ int main(int argc, char ** argv) {LOG_TEE("%s: warning: scaling RoPE frequency by %g.\n", __func__, params.rope_freq_scale);}- LOG_TEE("%s: build = %d (%s)\n", __func__, LLAMA_BUILD_NUMBER, LLAMA_COMMIT);- LOG_TEE("%s: built with %s for %s\n", __func__, LLAMA_COMPILER, LLAMA_BUILD_TARGET);+ print_build_info();if (params.seed == LLAMA_DEFAULT_SEED) {params.seed = time(NULL);diff --git a/llama.cpp/quantize/quantize.cpp b/llama.cpp/quantize/quantize.cppindex fce263b2..975d590f 100644--- a/llama.cpp/quantize/quantize.cpp+++ b/llama.cpp/quantize/quantize.cpp@@ -2,6 +2,7 @@// vi: set net ft=c++ ts=4 sts=4 sw=4 fenc=utf-8 :vi#include "llama.cpp/common.h"#include "llama.cpp/llama.h"+#include "llamafile/version.h"#include <cstdio>#include <cstring>@@ -94,6 +95,11 @@ static void usage(const char * executable) {}int main(int argc, char ** argv) {+ if (argc == 2 && !strcmp(argv[1], "--version")) {+ printf("llamafile v" LLAMAFILE_VERSION_STRING " quantize\n");+ exit(0);+ }+llamafile_check_cpu();if (argc < 3) {diff --git a/llama.cpp/server/server.cpp b/llama.cpp/server/server.cppindex 9919534b..01c0ca17 100644--- a/llama.cpp/server/server.cpp+++ b/llama.cpp/server/server.cpp@@ -10,6 +10,7 @@#include "llama.cpp/ggml-metal.h"#include "llama.cpp/ggml-cuda.h"#include "llamafile/llamafile.h"+#include "llamafile/version.h"#define CPPHTTPLIB_NO_EXCEPTIONS 1@@ -2628,6 +2629,11 @@ static void append_to_generated_text_from_generated_token_probs(llama_server_conint main(int argc, char **argv){+ if (argc == 2 && !strcmp(argv[1], "--version")) {+ printf("llamafile v" LLAMAFILE_VERSION_STRING " server\n");+ exit(0);+ }+llamafile_check_cpu();LoadZipArgs(&argc, &argv);diff --git a/llamafile/version.h b/llamafile/version.hnew file mode 100644index 00000000..54841db1--- /dev/null+++ b/llamafile/version.h@@ -0,0 +1,10 @@+#pragma once++#define LLAMAFILE_MAJOR 0+#define LLAMAFILE_MINOR 2+#define LLAMAFILE_PATCH 1+#define LLAMAFILE_VERSION (100000000 * LLAMAFILE_MAJOR + 1000000 * LLAMAFILE_MINOR + LLAMAFILE_PATCH)++#define MKVERSION__(x, y, z) #x "." #y "." #z+#define MKVERSION_(x, y, z) MKVERSION__(x, y, z)+#define LLAMAFILE_VERSION_STRING MKVERSION_(LLAMAFILE_MAJOR, LLAMAFILE_MINOR, LLAMAFILE_PATCH)