SanicAnima is fast Anima model.
The Full model is a checkpoint (AIO) model and the Pruned model is a diffusion model. If you know how to use the Anima model, use the Diffusion model. Otherwise, use the Checkpoint (AIO) model.
The recommended settings are: dpmpp_2m_sde_gpu / simple / CFG 2.5 / 15 Stpes
For detailed usage instructions, please refer to the metadata of the sample images and the Anima Readme document.
Description
v1
FAQ
Comments (7)
How did you make it faster exactly?
they didn't. they are just using fewer steps, which you can do with the base model too if you use quality/aesthetic style LoRAs.
I use a mix of NTRMix Style (0.25) and the 8step-Turbo LoRA (0.75) with dpmpp_2_sde/beta, CFG=2 and 15 steps, and it works really well.
@prozombie ah, I assumed it was some kind of Turbo training like ZiT, but that takes labs a lot of conpute to do. Would be neat to have something like that, does like a rigid model that's hard to fine tune completely though
what's the full model? integrated text encoder and vae?
Yes, it is an integrated model using ComfyUI.