Dall-E, Midjourney, Stable Diffusion (etc) - who's playing, and with which?
- Started 3 years ago
- Last post 3 months ago
- 938 Responses
- Nairn3
Ooh! The Image retexturing feature in the Midjourney editor's pretty much exactly what I've wanted from it for ages now. Neat!
- utopian0
- Generate images using one of the following models: Flux Pro, Flux Ultra, and Ideogram.utopian
- Generate videos using one of the following models: Kling, Luma, Hailuo Minimax, and Runway.utopian
- Enhance images using one of the following: Upscale, Scene Transfer. As well as training your own Lora models for custom avatars.utopian
- I seen this one, he says video generations take almost an hour each compared to using the actual tool that takes 5 minutes, no buenoYakuZoku
- Sounded good at firstYakuZoku
- You can uses all of these AI platforms within KREA starting at $10 per monthutopian
- but the first result he showed using all apps was a million time better than the final one with KREA_niko
- I tested Runway and Hailuo Minimax with same prompts as I did within KREA using their Runway & Hailuo model, the results were almost identical.utopian
- Ah good to knowYakuZoku
- milfhunter0
LOOOK what christmas has become
- AI slopgrafician
- i didn't think that ad could be any worse, but shitty ai made it so.
drek.hans_glib - https://i.imgur.com/…utopian
- https://www.youtube.…PhanLo
- not badYakuZoku
- PhanLo0
US Only at the moment
https://labs.google/fx/zh/tools/…
-
- plash2
More of Google Veo 2: Demo of Eating, Drinking generation
- utopian1
GPT writes perfect AI video prompts for you
Direct link to this GPT for free: https://chatgpt.com/g/g-AAeNQhq0…
- YakuZoku6
Made one of these things over the weekend, took about 12 hours to create, mostly used Halluo for the video generations, but it did block some of my generations like Marcellus Wallace with the ball gag in his mouth. I listed all the tools used in the description.
- Oops, Looks like I should update the thumbnail for embedsYakuZoku
- Are you doing each of these scenes as a separate video, then editing them together? Image to video for the initial frame? Or is this just YOLO prompt and hope?monNom
- Curious what workflow people are using for this as I haven’t played with much video stuff yet.monNom
- 1 clip at the timeYakuZoku
- I'm planning on making a tutorial on this when I get a few of these done and get my workflow downYakuZoku
- Cool. Reassuring there's still a human in the loop. Seems like you could even do a feature-length movie solo with the right workflow. (What a time to be alive!)monNom
- This is pretty insane. I can’t even get a 5 second clip in runway that doesn’t mutate into a freak and then I blow all my credits trying to correct it_niko
- yeah runway blows, i just killed my subscription, I still have a shitload of credits that i want to use up but Kling and Hailuo work so wellYakuZoku
- hans_glib2
it's hard to express how much i loathe so-called ai and the way it rips off any and everything it can find... absolutely disgusting and the end results are so bad.
however...
based on morilla's trailer below i'm having a blast having a crack at making the video we could never afford for a track we recorded long ago. i can even work with the endless cuts, enforced by the shitty tech.
hypocrite? moi?
- having said that the shitty thing can't understand a simple instruction to move through long grass at ground levelhans_glib
- also frankly WTAF???? https://hailuoai.vid…hans_glib
- Use Hailuo director mode for camera controlYakuZoku
- and omfg https://hailuoai.vid…hans_glib
- @yaku: i am, but shitty thing is shittyhans_glib
- prompt for thatnonsnese was "loaded shotgun is closed [track right, pan left] "hans_glib
- when Hailuo fails me i use Kling, both wouldnt do the ball gag on my vid below though, had to use Runway for that clipYakuZoku
- which surprised me because runway fails me 80% of the timeYakuZoku
- Kling is the shit though, and it exports @ 1080, for Hailuo i have to run it through an upscalerYakuZoku
- YakuZoku1
lol
- PhanLo2
some old SD style https://artificial-art.eu/
-
- PhanLo4
new image to camera move test model
https://huggingface.co/spaces/st…
-
