Distribute and run LLMs with a single file.
C++ cross-platformggufllama-cpplocal-ailocal-inferencelocal-llmopen-source-aisingle-file-executablespeech-to-text
Schlep in Chromium Zlib
Justine Tunney2232b0ef49ee
diff --git a/llamafile/cuda.c b/llamafile/cuda.cindex 9e1e57a6..b62c5e21 100644--- a/llamafile/cuda.c+++ b/llamafile/cuda.c@@ -51,6 +51,11 @@ __static_yoink("llama.cpp/ggml-common.h");__static_yoink("llama.cpp/ggml-backend.h");__static_yoink("llama.cpp/ggml-backend-impl.h");+// yoink the fastest zlib deflate impl from cosmo libc+__static_yoink("_Cz_inflateInit2");+__static_yoink("_Cz_inflate");+__static_yoink("_Cz_inflateEnd");+#define THESTRING(x) #x#define STRINGIFY(x) THESTRING(x)#define ARMS_ONLY(x) (!IsAarch64() ? "-DIGNORE" STRINGIFY(__COUNTER__) : x)