Distribute and run LLMs with a single file.
C++ cross-platformggufllama-cpplocal-ailocal-inferencelocal-llmopen-source-aisingle-file-executablespeech-to-text
Release llamafile v0.4.1
Justine Tunneyf6ea6bf54d16
diff --git a/.gitattributes b/.gitattributesdeleted file mode 100644index c44c4567..00000000--- a/.gitattributes+++ /dev/null@@ -1,2 +0,0 @@-# -*- conf -*--*.dll binarydiff --git a/README.md b/README.mdindex db20f945..9d0e3dd6 100644--- a/README.md+++ b/README.md@@ -152,7 +152,7 @@ enable you to work around Windows' 4GB executable file size limit.For Windows users, here's an example for the Mistral LLM:```sh-curl -L -o llamafile.exe https://github.com/Mozilla-Ocho/llamafile/releases/download/0.4/llamafile-server-0.4+curl -L -o llamafile.exe https://github.com/Mozilla-Ocho/llamafile/releases/download/0.4.1/llamafile-server-0.4.1curl -L -o mistral.gguf https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF/resolve/main/mistral-7b-instruct-v0.1.Q4_K_M.gguf.\llamafile.exe -m mistral.gguf```@@ -160,7 +160,7 @@ curl -L -o mistral.gguf https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1Here's the same example, but for macOS, Linux, and BSD users:```sh-curl -L https://github.com/Mozilla-Ocho/llamafile/releases/download/0.4/llamafile-server-0.4 >llamafile+curl -L https://github.com/Mozilla-Ocho/llamafile/releases/download/0.4.1/llamafile-server-0.4.1 >llamafilecurl -L https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF/resolve/main/mistral-7b-instruct-v0.1.Q4_K_M.gguf >mistral.ggufchmod +x llamafile./llamafile -m mistral.ggufdiff --git a/llamafile/cuda.bat b/llamafile/cuda.batindex dce31599..55d8c38c 100644--- a/llamafile/cuda.bat+++ b/llamafile/cuda.bat@@ -1,5 +1,3 @@-call "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvars64.bat"-nvcc -arch=all ^--shared ^--forward-unknown-to-host-compiler ^diff --git a/llamafile/version.h b/llamafile/version.hindex 6083fae6..468de843 100644--- a/llamafile/version.h+++ b/llamafile/version.h@@ -2,7 +2,7 @@#define LLAMAFILE_MAJOR 0#define LLAMAFILE_MINOR 4-#define LLAMAFILE_PATCH 0+#define LLAMAFILE_PATCH 1#define LLAMAFILE_VERSION (100000000 * LLAMAFILE_MAJOR + 1000000 * LLAMAFILE_MINOR + LLAMAFILE_PATCH)#define MKVERSION__(x, y, z) #x "." #y "." #z