Loras trained in Pixai.
Description
FAQ
Comments (4)
Illustrious is like a 'turbined/boosted' SDXL models, which is better (or a lot better) than PonyXL. I love poolrooms, this one will be great, I suggest a try in the future.
NoobAI is even better. NoobAI is basically just better Illustrious
Maybe Illustrious is better than Animagine 3, but NoobAI-XL (NAI-XL) is better than Pony Diffusion V6 XL and Illustrious.
I tell this because while Animagine and Illustrious can respond to prompts made with tags used in Danbooru only, Pony XL and NoobAI can respond to prompts made with tags used in Danbooru and/or e621, Pony has been trained with score tags (score_9, score_8_up, score_7_up, etc.), and Noob has been trained with more conventional tags (masterpiece, best quality, etc.), Noob understands more concepts than Pony, and Noob has the ability of emulate the style of a large amount of artists while Pony can emulate the style of a tiny amount of artists.
Loras/Doras/Lycoris/etc made for Illustrious are compatible Noob, and the ones made for Pony have somewhat compatibility with Noob, so I think there shouldn't be problems if a Noob based Lora is used with an Illustrious based checkpoint model.
I will keep using NoobAI for Lora training because I like that model more than Illustrious.
@CuauhtemocI5MAL I can confirm that using a noob lora with Illustrious models works great too. I use a mix and they almost always work, sometimes you need a VAE but otherwise it's good
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.