How to use James040/llama-cpp-python-wheels with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="James040/llama-cpp-python-wheels", filename="{{GGUF_FILE}}", )
output = llm( "Once upon a time,", max_tokens=512, echo=True ) print(output)
https://huggingface.co/James040/llama-cpp-python-wheels/resolve/d65ab05c5dea3e1b992e5e33dbb49edd1c9176a7/diskcache-5.6.3-py3-none-any.whl
hf download hf://James040/llama-cpp-python-wheels@d65ab05c5dea3e1b992e5e33dbb49edd1c9176a7/diskcache-5.6.3-py3-none-any.whl
curl -L -o diskcache-5.6.3-py3-none-any.whl https://huggingface.co/James040/llama-cpp-python-wheels/resolve/d65ab05c5dea3e1b992e5e33dbb49edd1c9176a7/diskcache-5.6.3-py3-none-any.whl
