[Feature] Load models from local files

[TamaMcGlinn]

Feature Request

I installed some models on one machine, but then decided I needed to switch to another one. I saw that on machine 1, the models I downloaded went into ~/.cache/gpt4all/ but after scp those to the other machine, it does not see any models installed, and clicking download still starts the whole download process. It would be great to have, besides download, a 'load from file' option that just transfers it on the local machine.

[ThiloteE]

You have been looking in the wrong directory. See https://github.com/nomic-ai/gpt4all/wiki/Frequently-Asked-Questions#where-are-the-default-directories-for-models-and-settings

In short: In GPT4All 3.9.0 they are here by default:

Windows

Settings directory: C:\Users\%USERNAME%\AppData\Roaming\nomic.ai
Models directory: C:\Users\%USERNAME%\AppData\Local\nomic.ai\GPT4All

Mac

Settings directory: /Users/{username}/.config/gpt4all.io
Models directory: /Users/{username}/Library/Application Support/nomic.ai/GPT4All

Linux

Settings directory: /home/{username}/.config/nomic.ai
Models directory: /home/{username}/.local/share/nomic.ai/GPT4All

But you also can configure the model directory in the GPT4All-Chat settings.

From those directories, you can copy or move those models to somewhere else.

[ThiloteE]

Sideloading, as we call it requires a restart of GPT4All. GPT4all once had the ability to automatically detect new models, but that was removed for reasons. I don't remember the reasons, but maybe it made the code base too complicated.