absolutely great!
I have tried a lot of gemma-4-31b-it finetunes, gemma-4 being my current favourite. And while I have limited experience with this one yet, the model card does not lie: minimal damage, and great writing style. Pretty sure this is my current go-to model now. At least, it made me make this semi-excited comment :) Seems to be a great method. Thanks!
I haven't tried much 31B models myself, the speed penalty is enough where i prefer 26B models (generally to RP). Actually prefer MOE models in general after seeing the speed boost if that's how they are going to perform.
Still, might be worth a try.
yeah, gemma-4-26b-a4b is a speed monster even on pure cpu, or with very little vram. i wonder if a gemma-4-26b-a4b-styletune is in the cards, it has even more need for it than the 31b :)
It wasn't in the cards until a minute ago. 😉 I'll see what I can do!
Just a humble note. I'm currently testing this with Gemma4-12b-it and it seems to be working very well, too. Many thanks!
Great to hear! I love how fast 26B-A4B is to train out of the box, so https://huggingface.co/Gryphe/Gemma-4-26B-A4B-StyleTune is a thing now. Enjoy!
Gemma-4-26B-A4B-StyleTune is a thing now. Enjoy!
Woo! Once it's GGUF'd i'll be downloading it. I'm sure it's already in the pipeline for @mradermacher .
Absolute cinema! I hope you'll release version 12b too.
wow, heroic overnight delivery :)
And from what I can see, it's a success. The 26B probably benefits even more than the 31B, style-wise. But damn, the 26B is much more stubborn with refusals.
And from what I can see, it's a success. The 26B probably benefits even more than the 31B, style-wise. But damn, the 26B is much more stubborn with refusals.
Then should it have been heresy/uncensored patched before finetuning with the style? Or go the other way around?
One tensor changed out of 659.
Hmmm which tensor was it i wonder? By splitting the individual layers out i could swap it with an ARA heresy one i have enjoyed and see how that fares.
It's really exciting to see these results. +beer to Gryphe.
Wow! My go-to for my SillyTavern roleplay has been SkyFall for months now (specifically TheDrummer_Skyfall-31B-v4.2-Q5_K_M - I'm on a laptop 5090, 24GB VRAM). I do mostly interactive stories with rich lorebook worlds. But just like that - StyleTune is a drop in, and just works. THANK YOU for this. It's still early days, and I'm still using mostly settings from prior gemma fails.
Please make a Heretic version @mradermacher @llmfan46 My two inspirations to pursue AI!
I am very busy working on quite a few other AI projects at once right now, but I MIGHT be able to have the uncensored version of this model released in a week or two though.
Please make a Heretic version @mradermacher @llmfan46 My two inspirations to pursue AI!
I am very busy working on quite a few other AI projects at once right now, but I MIGHT be able to have the uncensored version of this model released in a week or two though.
I just realized, this is board not for the MeroMero styleswap 31B Version, my mistake.
Also, @llmfan46 You're the inspiration I abliterated my first model! Thank you!
Also, @llmfan46 You're the inspiration I abliterated my first model! Thank you!
Your're welcome, congratulations and thank you.
I just realized, this is board not for the MeroMero styleswap 31B Version, my mistake.
Happens. I remember leaving a note suggesting a model was really good, only to realize it was not the same model at all... quickly closed the discussion as it was invalid at that point.
Easy to get lost when half the pages all look/feel the same, and model names get blurred together.