M1 Install Failure

[imaurer]

I would appreciate any tips or tricks for figuring out the next step. I get the same error with brew python or miniconda.

Below is some of the information I think is most relevant.

% python -c 'import platform; print(platform.machine())'
arm64

% python -V
Python 3.10.11

When I try to install:

% CMAKE_ARGS='-DLLAMA_METAL=on' FORCE_CMAKE=1 \
  pip install llama-cpp-python --force-reinstall --upgrade --no-cache-dir
Collecting llama-cpp-python
  Downloading llama_cpp_python-0.1.77.tar.gz (1.6 MB)
...
      [5/8] Linking C static library vendor/llama.cpp/libggml_static.a
      [6/8] Linking CXX shared library vendor/llama.cpp/libllama.dylib
      FAILED: vendor/llama.cpp/libllama.dylib
...
      Undefined symbols for architecture arm64:
        "std::__1::basic_stringbuf<char, std::__1::char_traits<char>, std::__1::allocator<char>>::str() const", referenced from:
            _llama_copy_state_data in llama.cpp.o
        "std::__1::basic_filebuf<char, std::__1::char_traits<char>>::open(char const*, unsigned int)", referenced from:
            std::__1::basic_ifstream<char, std::__1::char_traits<char>>::basic_ifstream(char const*, unsigned int) in llama.cpp.o
        "std::__1::basic_filebuf<char, std::__1::char_traits<char>>::basic_filebuf()", referenced from:
            std::__1::basic_ifstream<char, std::__1::char_traits<char>>::basic_ifstream(char const*, unsigned int) in llama.cpp.o
        "std::__1::basic_filebuf<char, std::__1::char_traits<char>>::~basic_filebuf()", referenced from:
            llama_apply_lora_from_file_internal(llama_model const&, char const*, char const*, int) in llama.cpp.o
            std::__1::basic_ifstream<char, std::__1::char_traits<char>>::basic_ifstream(char const*, unsigned int) in llama.cpp.o
            std::__1::basic_ifstream<char, std::__1::char_traits<char>>::~basic_ifstream() in llama.cpp.o
   ....
        NOTE: a missing vtable usually means the first non-inline virtual member function has no definition.
        "vtable for std::__1::basic_stringstream<char, std::__1::char_traits<char>, std::__1::allocator<char>>", referenced from:
            std::__1::basic_stringstream<char, std::__1::char_traits<char>, std::__1::allocator<char>>::basic_stringstream[abi:v15006]() in llama.cpp.o
        NOTE: a missing vtable usually means the first non-inline virtual member function has no definition.
      ld: symbol(s) not found for architecture arm64
      clang: error: linker command failed with exit code 1 (use -v to see invocation)
      [7/8] Linking C shared library vendor/llama.cpp/libggml_shared.dylib
      ninja: build stopped: subcommand failed.

The full error log is here:
https://gist.github.com/imaurer/668f61b4cb71fb85fedd6cadf90072c1