I may decide to stop doing all 9 models in these polls when i request this much text. I have done enough requests to know it's usually Recraft, Ideogram, Pro then Ultra then Dev ... that's the order of accuracy. So this may be one of the last where i use all 9 models when i have this much text. BUT running these blind tests are still great to do so it's not just matter of theory but becomes tested.
I think that maybe next time i'll run the Recraft/Ideo/Pro and then test dev and ultra and see how well they do. Maybe i'll add 2 of each model and see who comes out on top.
USE https://peakd.com/ai and run your own BLIND POLL and use #ai-images
P.s. There is a competition use SNAPS and share some posts 1. Go here: https://peakd.com/c/hive-163772/snaps 2. Write a snap (go see other examples) 3. Use ALL 3 of these #worldmappin #winterchallenge #traveldigest Limit 2 entries a day You can use any of the images from below... but of course you know the prompt to use to make your own. haha
THE PROMPT
A promotional image for a contest that is all about the world and maps with text that says "Worldmappin Winter Challenge" and then also says "#worldmappin"
Choose top 3 images that reflect the PROMPT above
THE IMAGES
IMAGE 1 (Below)
IMAGE 2 (Below)
IMAGE 3 (Below)
IMAGE 4 (Below)
IMAGE 5 (Below)
IMAGE 6 (Below)
IMAGE 7 (Below)
IMAGE 8 (Below)
IMAGE 9 (Below)
THE METHEDOLOGY
STEP 1. Ran each model twice with Same Aspect ratio
NOTE: Schell gives 3 images at a time and DEV gives 2 by default
STEP 2. I Chose the best image (Favoring correct spelling) STEP 3. I added in midjourney 1 run = 4 images (standard 6.1, stylization 300, 0 weirdness, 10 variety)
YOU NOW VOTE
- Pick your top 3 results in the poll
- I won't share the models right now...
- Later I'll share which models ranked highest
The cost of the test is 0.97HBD (does not include midjourney subscription)
Published with PeakVault
