Flux
I Pomo Mix [Ipomo + NAI + Pony + Artists Etc.] V1.0
model_id: i-pomo-mix-ipomo-nai-pony-artists-etc-v1-0
This Lora was trained on I_Pomo + a mix of NAI images to get that NAI style that we all like, and on my own stuff based on Pony to give balance. The dataset was carefully selected to be unconventional but not too crazy.Some nice features of this Lora:-Eye enhancement in done in every single image on the dataset, even in the original artist pictures. So most sizes and distances get the eyes right and you don't need to upscale. Nice pictures with no upscale from several distances.-A nice little first step in the direction of a NAI type style using Flux-Hands and eyes corrected on the dataset that contained Pony + puuzaki puuna and other artists.Trained on a H100, batch size 2, 3000 steps, using Ai-Toolkit.About Lora scale: The sweet spot is between 0.6 to 0.8. If you go to 1.0 to 1.2 you can get awesome NAI style pictures but text will be ruined. Hands and feet will also suffer. 0.6 at 28 steps is a nice spot. 0.8 at 25 steps is my sweet spot right now. I'm still testing it out. I still haven't tested sigmas etc.
NSFW samples are gated
This model produces NSFW content. Log in and enable NSFW in your account settings to view samples and run the model. The API and model parameters below are accessible regardless.
Use I Pomo Mix [Ipomo + NAI + Pony + Artists Etc.] V1.0 via the ModelsLab API
Send a POST request to the relevant endpoint with your API key and model_id="i-pomo-mix-ipomo-nai-pony-artists-etc-v1-0". The API does not enforce the same NSFW gating as the web playground — content moderation is handled by your account-level policy. Sign up to get an API key.