Yep. Using plus. Think it’s £19 a month or somethingdo most people here use the paid version of chatgpt? signed up last night to play with the image generation, only 2 or 3 images allowed per day on the free sub
Midjourney - AI art
- Thread starter tang0
- Start date
I don't believe you actually highlighted a limitation of the model though, you just made an assumption that didn't actually hold up when tested.
do most people here use the paid version of chatgpt? signed up last night to play with the image generation, only 2 or 3 images allowed per day on the free sub
Yes they're getting absolutely hammed by the popularity of these things right now, the model is really good (despite misplaced skepticism), Sam Altman has commented a few times on twitter or X about staff working late etc.. and the roll out to free users was initially delayed.
You need to understand why I used the specific wording I did and put it in quotes or otherwise you'll just continue with comments like this which entirely miss the meaning of what I was saying.
do most people here use the paid version of chatgpt? signed up last night to play with the image generation, only 2 or 3 images allowed per day on the free sub
There are quite a few different models (or variations of) out there and varying products/prices with different strengths and weaknesses - mostly unless I have specific needs I either use DALL-E based or a couple of apps designed for video game asset generation which use derivatives of Stable Diffusion.
EDIT: On that note I need to check out how the 4o update handles seamless tiling materials - something most of the models don't really understand - some have a tile command but it is usually a post processing layer which doesn't always produce great results.
Last edited:
If you think there's something being missed then just say what it is, I think you're mistaken though - this sort of thing can be tested and I don't think it's helpful to spread misinformation.
------------------------------------------------------------------
The new model is very good and I think it's worth treating claims about limitations with a lot of skepticism - I'd not necessarily assume misunderstanding re: the model either on the basis that you didn't get a result you wanted - on the topic of limitations or model understanding here is a funny one:
Ask it to draw Elon - it does it right away:
Now in a different chat session ask it to identify the person in the image it just created - the first response might make someone assume that it can't do that, but of course, it's perfectly able to it's just forbidden to do so:
Last edited:
Associate
- Joined
- 13 Sep 2010
- Posts
- 2,455
Is it me, or does the boat look like it's stretching, more than moving?
Looking at the ground more than the Liner, does for me give an impression of moving.
Looking at the Liner only does give a slight impression of stretch imo.