New release of llm-gpt4all - it's now really easy to run models like mistral-7b-instruct-v0 on your own machine:

pip install llm
llm install llm-gpt4all
llm -m mistral-7b-instruct-v0 "ten facts about pelicans"

https://github.com/simonw/llm-gpt4all/releases/tag/0.2

Release 0.2 · simonw/llm-gpt4all

Upgrade for compatibility with the latest gpt4all. #16 The new version of gpt4all uses models in the gguf format. You may have existing ggml models - these will no longer be used by the plugin, so...

GitHub

If you've installed a previous version of the plugin you can upgrade it like this:

llm install -U llm-gpt4all

More about the underlying GPTAll project here https://gpt4all.io/index.html

GPT4All

Free, local and privacy-aware chatbots

@simon Thanks! Is there an image processing equivalent. One where I can run queries against a photo or compare two photos? I did look at the models on the LLM site
@jotkailay I have not got LLaVA working through LLM yet but that's an open model that can to images as well as text - you can play with that one online here https://llava-vl.github.io
LLaVA

Visual Instruction Tuning