ComfyUI AI: SDXL Base Model only, creative testing with image to image + HiRes fix.
ฝัง
- เผยแพร่เมื่อ 9 ส.ค. 2023
- With a simple setup in ComfyUI AI, I tested the SDXL Base model together with Image to Image and HiRes-Fix in terms of its creative use. Even without the provided SDXL refiner model, the results in the higher default resolution of 1024x1024 are already promising. In my opinion, it will be able to compete fiercely with Midjourney in combination with the hopefully soon to be released Controlnet models. I am already quite excited!
- วิทยาศาสตร์และเทคโนโลยี
Deutsch erwähnt, gutes video 😂👍
Aber klar doch! ;) Danke!
For what it is worth, I was fully able to follow along with this video while keeping the audio muted since it was too distracting for me.
I do not mind videos that have no voice, as long as the text is clearly readable and a few annotations are added along the way.
This video did that quite well.
Great!! Subscribing you!!👍
Thank you, I sub second
👋
You need to rethink your method of explaining things .Most people including myself like to hear explanations and not music and sound effects .The reason is that not everyone likes the same music so you lose a great proportion of potential viewers .It looks like you know what you are doing so it's a shame that chose this way .Best of luck .
Hey, thanks for the tip and for the compliment. I'm still trying to figure out what works best for me regarding video creation. Recording the voice is still my Achilles heel, maybe I'll give it another try anyway. Have a pleasant day!
Have faith .@@Showdonttell-hq1dk
hahaha yeah, probably easier to load your text into a text to voice ai generator if you don't wanna speak my guy!
@@CPDCT Yes, maybe a good idea, I'm trying it out at the moment. I tested it a few months ago, but the results weren't that convincing. In the meantime, the AI models have improved. Thanks for the reminder!