Hacker Newsnew | past | comments | ask | show | jobs | submit | kornesh's commentslogin

Yeah, solving the flickering problem and achieving temporal consistency will be the key to realize the full potential of generative video models.

Right now, AnimateDiff is leading the way in consistency but I'm really excited to see what people will do with this new model.


Thank you! I spent a lot of time reading reviews of both options and ultimately decided to go with Lexical because its API seemed easier. However, I quickly realized that the API is kind of twisted and has many restrictions. Also, the documentation was sparse but by then it was too late to restart from scratch.

Do you have any experience between these two?


I'm over the the moon! absolutely wasn't expecting a PR, let alone two! Thanks @smcleod :)


Hi, I've spent past couple of weeks working on this. I realize there are many of them now but there was none when I first started working on this. The pace at which this community is putting out new tools is crazy. Also, personally this is first thing that I've build so far that I spent hours using lol, so I hope some else finds it cool. Wasn't really sure if I should post here, but here it is. Roast me!


I think that lexica.art has the prompt search space mostly cornered now, but I think there might be room for a tool that aggregates various styles together (greg rutkowski, van gogh, roy lichenstien, etc etc) and shows their overall popularity? Could pivot in that direction?


There's already a site for that, let me see if I can find it again. The open source community is going all out on everything stable diffusion!

Edit: Do you mean something like this https://f000.backblazeb2.com/file/clip-artists/index.html


Yeah, that'll work. Doesn't look as good as your page. Maybe a StumbleUpon for art styles? (God, I miss StumbleUpon)


I'm working on building a search engine specifically to help with prompt engineering https://pagebrain.ai/promptsearch/ Is this useful enough to be a startup?


Where are you scraping data from?

Appears super similar to Lexica.art, but less data; for example, appears you’re missing even the limited about of parameters listed by Lexica, which still are incomplete.

https://lexica.art/


From stable diffusion discord channel. That looks much better!


Yep, same place they’re getting it from; exchanged messages with them too.


That's cool! Please do apply.


Yes.


Google Cloud Storage can host static sites and has a free forever tier


Reminds me of Elon's quote "product errors reflect organizational errors"


I thought I was the only one to get offended every time I have to relogin lol. AWS is the absolute worst, logs me out every 5 minutes.

I have been fascinated about google's identity aware proxy design. We have adopted something similar at much smaller scale with kubernetes and sidecars, I can attest we're absolutely more productive.


On the point about AWS, there are settings in your account to prevent that happening - whoever is managing it may have set the log out time a bit low.


Even then the max session is 12 hours which can be cumbersome, though we've made it work using an IdP into AWS SSO.

12 hours is typically enough but, for the days it isn't (i.e. something breaks and I'm working over time), it becomes tiresome.


How does residential IPs actual work?


TorGuard's "Streaming IP" certainly does. It allows me to use streaming services and sites that are normally blocked for me at my home in Mexico.

I just route my entire connection through it on one of my SSIDs/VLANs.


You filter by block's AS. It's not always 100% correct, but with enough people you can collect your own list of most common VPS / hosting companies.


Maybe ISPs have a list of IPs to assign to residential customers?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: