TL;DR: GPT4ALL
Source:
GitHub - nomic-ai/gpt4all: gpt4all: an ecosystem of open-source chatbots trained on a massive collections of clean assistant data including code, stories and dialogue
gpt4all: an ecosystem of open-source chatbots trained on a massive collections of clean assistant data including code, stories and dialogue - GitHub - nomic-ai/gpt4all: gpt4all: an ecosystem of ope…
What ?
GPT4All is an open-source chatbot based on LLama, that can run on less powerful PCs than the war machines commonly used for AI. (i7 - 8Go RAM will do slowly, but surely). Trained model of 7 Billion parameters, so don't expect complex answers. Dev, trained & fine-tuned by Nomic-AI.
How to :
- Download the model :
gpt4all-lora-quantized.bin (or Torrent)
or
gpt4all-lora-unfiltered-quantized.bin (or Torrent)
- Clone the repo
git clone https://github.com/nomic-ai/gpt4all.git
- Move the downloaded model in the /chat/ folder
- Run the app in a terminal
gpt4all-lora-quantized-win64.exe
- For the unfiltered version :
gpt4all-lora-quantized-win64.exe -m gpt4all-lora-unfiltered-quantized.bin