Instructions to use TheBloke/mpt-30B-chat-GGML with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use TheBloke/mpt-30B-chat-GGML with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("TheBloke/mpt-30B-chat-GGML", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Why not llama.cpp ?
#5
by lazyiitian - opened
Why can't this run on llama.cpp when it can run on others that use llama.cpp?
What's missing?
llama.cpp only supports Llama models. This is an MPT model - different model type, different code required to load it.
It's hoped that some time in the future - maybe in a month or so - llama.cpp will support more model types than just Llama, including MPT and Falcon. But for now, it doesn't.
But you can use other clients for this, like koboldcpp, LoLLMS-UI, ctransformers, etc.