Instructions to use TheBloke/Falcon-7B-Instruct-GGML with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use TheBloke/Falcon-7B-Instruct-GGML with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("TheBloke/Falcon-7B-Instruct-GGML", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Text Generation WebUI says it supports GGML models?
@TheBloke why you say GGML model don't work in text-generation-webui?
from https://github.com/oobabooga/text-generation-webui#ggml-models
"You can drop these directly into the models/ folder, making sure that the file name contains ggml somewhere and ends in .bin."
I don't. I say THESE GGML models don't work in text-generation-webui.
text-generation-webui's GGML support is provided by llama-cpp-python, which currently only supports the same GGML models as llama.cpp, which means Llama and OpenLlama models. Not Falcon, MPT, GPT-J, GPT-NeoX, StarCoder.
These Falcon models are using a particular GGML library - actually called GGCC now - which is currently not supported by any UI, but hopefully this will change soon.