Download
1 variant available
fp16 SafeTensor
splatoonall-000010.safetensors
Half precision, best balance • 144.11 MB
Verified: 3 years ago
5910 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
1.9K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
1250 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
Mar 12, 2023
Same rules as the main Combined, which is e8. Except this one is e10.
Both of these give good results, but they're a little bit different. You'll probably like e8 better, but if you're a data hoarder you can grab this too.
Show more

1180 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
1.9K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
this is an exceptionally old model that i made a very long time ago. i have no plans on updating these, modern models can do these characters on their own just fine. nonetheless, this will stay up. the following description remains unmodified
---
NEWEST: ACTUALLY COMPETENT TAGGING. ALL EXAMPLE IMAGES ON COMBINED V2 ARE RAW GENS
THESE ARE LORAS FOR THE MAIN SIX SPLATOON CHARACTERS: CALLIE, MARIE, PEARL, MARINA, SHIVER, AND FRYE.
THESE ARE NOT LORAS FOR GENERATING INKLINGS OR OCTOLINGS IN GENERAL. MODELS LIKE AOM3 CAN ALREADY DO THAT JUST FINE.
NEW: Combined LoRA. this is the recommended version to use unless you know what you're doing.
If you're having problems getting a certain character to generate properly(this is most noticeable with shiver), try not including a species tag.
Trained on NAI
(Also keeping up the individual LoRAs. Wouldn't recommend using them, but hey, I'm not your dad)
\/-------------------- OLD DESCRIPTION --------------------\/
6 LoRAs, 6 characters.
Pearl, Marina, Callie, Marie, Shiver, Frye
Marina and Shiver are trained on AOM2-sfw
The rest are trained on NAI
All attached LoRAs are epoch 6, and all example images were made with 0.8 LoRA weight

