chore: roll back to lama_cpp 0.2.76

Makefile

@@ -16,7 +16,7 @@ install_cuda:
 	poetry config virtualenvs.in-project true
 	poetry install --extras "cuda-acceleration" --no-root --no-ansi
 	echo "Installing llama-cpp-python and ctransformers with pip to get NVIDIA CUDA acceleration"
-	. .venv/bin/activate && CMAKE_ARGS="-DGGML_CUDA=on" pip3 install llama-cpp-python==$(llama_cpp_version)
+	. .venv/bin/activate && CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip3 install llama-cpp-python==$(llama_cpp_version)
 	. .venv/bin/activate && pip3 install ctransformers[cuda]==$(ctransformers_version)
 
 install_metal:
@@ -25,7 +25,7 @@ install_metal:
 	poetry config virtualenvs.in-project true
 	poetry install --no-root --no-ansi
 	echo "Installing llama-cpp-python and ctransformers with pip to get Metal GPU acceleration for macOS systems only (it doesn't install CUDA dependencies)"
-	. .venv/bin/activate && CMAKE_ARGS="-DGGML_METAL=on" pip3 install llama-cpp-python==$(llama_cpp_version)
+	. .venv/bin/activate && CMAKE_ARGS="-DLLAMA_METAL" pip3 install llama-cpp-python==$(llama_cpp_version)
 	. .venv/bin/activate && CT_METAL=1 pip install ctransformers==$(ctransformers_version) --no-binary ctransformers
 
 install_pre_commit: