Z-Image Lora training test, Used for Z-Image aesthetic enhancement, improving illustration and artistic concepts.
SeaArt is now usable. cfg cannot be adjusted, but there's not much loss of data.
Recommended settings: heunpp2 + linear_quadratic, cfg: 1.5, step: 20, 0.8-1.2
The LoRa dataset uses a variety of training sets, which affects various styles, including realism, illustration, and animation. It also fixes the aesthetic issues of Z-image. Because LoRa training disrupts the acceleration effect, it requires 20 steps and cfg 1.5-2.0 to be effective. You can try the others yourself.
Due to the difficulty of Z-Image training, the step count and CFG need to be increased. I will update this when a better training method is available in future Base versions. Have fun!
Description
heunpp2+linear_quadratic，cfg：1.5，step：20，0.8-1.2
The LoRa dataset uses a variety of training sets, which affects various styles, including realism, illustration, and animation. It also fixes the aesthetic issues of Z-image. Because LoRa training disrupts the acceleration effect, it requires 20 steps and cfg 1.5-2.0 to be effective. You can try the others yourself.