AkashicPulse EQVAE is a model based on AkashicPulse series, but fine-tuned to utilize Anzhc's EQVAE as its VAE. (In theory,) This improve the model generation color and accuracy.
Recommended settings [UPDATED FOR V1.0.x]:
Same like v0.2, the model is in between needing RescaleCFG and not needing it. Expect to turn on and off this a lot.
The usage of artist tags IS HIGHLY RECOMMENDED. Put it at the FRONT of your prompt (and use BREAK if your generation platform support it).
Steps: 32-50
CFG: 7-10 with RescaleCFG on or off, or 4.5-6 if you don't wanna touch RescaleCFG
RescaleCFG: 0.7 OR OFF, inspect generated image quality if something is off.
Sampler: Euler a (or try AkashicSolver available on Adept Sampler with AkashicAOS).
Noise Scale: 0.9, using 1.0 will cause the model to over-generate noise.
Recommended negative prompt:
worst quality, low quality, bad anatomy, bad hands, text, signature, watermark, username, artist name, [things you want to remove]WARNING: Using excessive negative prompt might resulting the model to lose its quality.
Model Specs:
v1.0.2
Based on: AkashicPulse EQVAE v1.0.1-Base
Dataset size: ~21k
Epoch posted: Epoch 10 out of 10
EQVAE used: Anzhc's EQVAE B7
v1.0-v1.0.1
Based on: AkashicPulse EQVAE v0.2
Dataset size: ~4.6k
Epoch posted: Epoch 15 out of 15
EQVAE used: Anzhc's EQVAE B7
v0.2 [CHECK ABOUT THIS VERSION FOR DETAILS]
Based on: AkashicPulse EQVAE v0.1
Dataset size: ~9.4k
Epoch posted: Epoch 7 out of 10.
EQVAE used: Anzhc's EQVAE B7
v0.1
Based on: AkashicPulse 4.1
Dataset size: 10.6k, repeated twice.
Epoch posted: Epoch 9 out of 10.
EQVAE used: Anzhc's EQVAE B7
Description
I think the model is ready to be called a v1.0. It has superior quality over v0.2 and able to generate amazing results.
Same like v0.2, the model is in between needing RescaleCFG and not needing it. Expect to turn on and off this a lot.
FAQ
Comments (5)
Hi, it is recommended to use the RescaleCFG extension as I mention on the model description. It will tame down the lighting, while keeping the detail of high CFG.
IIRC it is built-in to ComfyUI and SD-Webui reForge. Not sure with SD-Webui Forge.
@kayfahaarukku Hi, I didn't know anything about it, now I found the option, I activated it, and let's see if it generates it now without blinding me XDDD
I like its potential, but it is too hard to use for the results it gives, at least for me.
Thank you for the feedback. In the meantime, I'm still refining my methods on how the model should be trained so it works without changing too many parameters when generating, since usually EQ-VAE is paired with V-prediction models, and this model is Epsilon prediction. There's a gap of knowledge that I need to fill myself.
Please enjoy observing the slow-but-sure process for now 🙏