Character: Hitori Gotou
Anime: Bocchi The Rock
Recommended model: Waifu Diffusion 1.4+
Follow us on Pixiv and see our full sets: https://www.pixiv.net/en/users/93270522
Donations (Will help us a lot): https://shorturl.at/ejJLT
NOTE: Is recommended to add extra quality layer to improve eye quality with https://github.com/hako-mikan/sd-webui-regional-prompter.
Add base prompt (A.K.A ADDBASE) and past in both addbase and specific prompt the same content (except LoRA activation tag) and you will get an extra quality layer.
Description
FAQ
Comments (15)
Hello, can you provide the prompt for the example pictures?
When you make a lora set. The prompt is so that people can know the trigger word
Your lora doesn't works, what should I do? I also get "The size of tensor a (768) must match the size of tensor b (1024)" error.
We have checked that error in google. It seems you are trying to use that model with 1.5 ControlNet or any 1.5 textual inversion / LoRA. Try to check if they are all in 2.1. If you find something else send us a comment :)
Link of the model/checkpoint you are using?
@Furuhata Waifu Diffusion 1.4
@TheHolySee I tried using that model but I don't know if is the correct one, I don't get the error anymore, but I can't use your lora or it generates a random image, can you help me please :(
@Furuhata Write us on Twitter and show us your SD setup and we’ll try to help you
@TheHolySee There is no pm option in your twitter :(
@Furuhata Done. Write
@Furuhata we are waiting for you if still needing help. Also if you like you can send your Twitter/Discord user and we'll write to you.
@TheHolySee I still can't sent you a pm :/
@TheHolySee Just pm'd you on your pixiv, check please.
ぼっちちゃんかわよ
Sorry but I have to say that while other loras run well but only your lora comes up with the error. Maybe you can fix it or else people can't use your work. Same error with other loras you upload, like Ikuyo Kita.
RuntimeError: The size of tensor a (768) must match the size of tensor b (1024) at non-singleton dimension 1Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.


