Gryphe, creator of MythoMax basically merged the best Mistral models together. This should be a really fantastic model!
https://huggingface.co/Gryphe/MythoMist-7b (links to quantized models by TheBloke can be found there)
Edit: oof, messed up the title. It’s MythoMist, not Mix.
You must log in or register to comment.
links to quantized models by TheBloke can be found there
Look, I know we’re all incredibly grateful for the work TheBloke puts in, but seriously…does literally no one else make these? Like it just seems crazy for one person to be holding it down like this.
I make some exllamav2 ones (including mythomist) as well here for anyone interested: https://huggingface.co/bartowski
Undi does GGUFs of his own models, they’re pretty great.
There is no real logic in how these models were divided throughout the merge
I’m kind of cautious how random merging affects the overall quality, since many of these merges models were trained with different prompt formats. In my experience that would inevitably lead to AI outputs that attempt some gibberish by adding bits of other used prompt formats (e.g. “### Response:” being printed out while using the ChatML template). To my surprise I witnessed that with OpenHermes 2.5 in some edge cases. But I would be eager to hear other people’s experience on this.
+1 on the fact that I have infrequently seen my own merge experiments spit out “### Reponse:” while using the standard Alpaca format. It so far hasn’t happened enough times to be a concern. I’ve seen more benefits than deficits when it comes to merging models overall however. I’m coming at this primarily from a creative writing aspect, so I’ll usually find models that have characteristics I enjoy, merge them together, and then continue merging models down the line to experiment with output quality.
How can I use this in lm studio?
Damn, no 13B?
It’s a mistral base model which only exists in 7b variety.
Mistral is beloved for having 13b quality in the size and speed of a 7b model.
7b models are much better than most 13b models at the moment. Maybe when Mistral 13b comes out you’ll see more of them.