Eight Buffalo Media Group has shared our Newest SD 1.5 8Buff Mix text-to-image model!

Eight Buffalo Media Group has shared our Newest SD text-to-image model!

Model Details

Our latest 8Buff_Mix model is a mix of our 8Buff_Gen model (https://civitai.com/models/41928/8buffgen) and 8Buff_Real (https://civitai.com/models/118761/8buffreal?modelVersionId=253177) SD 1.5 models, with some additional training and adjustments for improved hands and prompts. Note that Version 1 of this model still has a few issues, so please be patient as we improve. Constructive feedback is always welcome.

Model Description

This model does require a good deal of prompt crafting, the trade off is you have a lot of control on the images you create. It has a focus on multiple people and of different looks and ages, with prompts of elderly or 60_years_old. It will also do nationality or skin color. Very stable with images for short video and GIFS (checkout our how-to here: https://eightbuff.blogspot.com/2023/06/article-seven-my-text-to-img-journey.html).

Technical Details

Works well for 512x512, 512x640, and 512x768 as well as combinations of those dimensions.
Preferred Sampling methods are; DDIM, DPM++ 2S a Karras, Restart, DPM++ 2M SDE Karras, and DPM++ 3M SDE Karras.
CFG Scale ~7
Works well with many LORAs.

Available on civitai.com : https://civitai.com/models/479886?modelVersionId=533718

Also available on Hugging Face : https://huggingface.co/EightBuff/Mix