İlker
@ilkerigz
Full Stack Developer with UI/UX Expertise
I created a Flux Kontext LoRA using @FAL that transforms your logos or objects into 3D metallic visuals. You can try it out or download the LoRA via the link in my next tweet. Very soon, I’ll launch a site where you can find both my LoRAs and ones shared by other creators.
it's true... i'm back again with a new LIMITED EDITION feature for ai memes MAGIC MERGE - this tool let's you lazily slap multiple pictures together like it's nothing! run to ai memes dot com to try it yourself (free for everyone) and make motha fuckin noise in the replies
yo i just pushed live yet another feature to the growing list of tools available on ai memes Magic Erase - this tool can blend stuff, it can remove stuff and what the hell... it can even add stuff free for everyone... please screech
ceku balım hazırsan çıkalım @FAL etkinliğinden sonra ve @ilkerigz 'in yaptıklarından sonra karar verdim, astria'da fal'a geçiyorum hem hız hem api kalitesi hem destek ile fal'ın daha iyi olduğuna karar verdim ama değiştirmek istemediğim için geçmiyordum. şimdi geçiyorum :D
🔥🔥🔥
Some news: We're building the next big thing — the first-ever AI-only social video app, built on a highly expressive human video model. Over the past few weeks, we’ve been testing it in private beta. Now, we’re opening early access: download the iOS app to join the waitlist, or…
Reminded me of these kinds of videos, and gave it a quick try. fal.ai/models/fal-ai/…
Having fun with @fal's LTX 0.98 Multicondition endpoint to make more music backgrounds. Viliam Lane's "Rei" here.
Deploying on Supabase Cloud & Vercel = handing them 50% of your company lol At least self-host Supabase, and throw Coolify on Hetzner and keep your stuff yours. Been doing this for 3-4 yrs, way better.
If this isn't your tech stack in 2025, you are ngmi
I’ve seen a lot of people mention (and I’ve noticed it in my own DMs too) that using a really big dataset with Kontext can actually make things messier. Honestly, same here. I’ve tried training a bunch of LoRAs with Flux Kontext and hit those issues myself. But here’s what I’ve…
Cooking up my biggest Kontext LoRA set yet (1500 pairs, thanks to @FAL ) with a pretty WILD (really WILD) captioning technique. If this works, it could be the craziest price-performance try-on model out there dirt cheap and insanely advanced. Wish me luck it all runs smooth
Cooking up my biggest Kontext LoRA set yet (1500 pairs, thanks to @FAL ) with a pretty WILD (really WILD) captioning technique. If this works, it could be the craziest price-performance try-on model out there dirt cheap and insanely advanced. Wish me luck it all runs smooth
Coding with AI can be more draining than just doing it the old-fashioned way sometimes. How does the exact same model in the same setup go from genius to straight-up clueless within the same day?
Trained a Tattoo LoRA for fun today, used just 25 pics and it’s already matching or beating most tattoo apps out there. Plus, it’s 10x cheaper. WILD
Had the chance to meet the amazing @FAL team and fellow AI enthusiasts today at the Istanbul meetup. Thanks for the great event!
For anyone wondering: It doesn’t really work well with couple photos. Or at least, not for some of them in my experience…
Trained a FLUX Kontext LoRA that makes your photos look way more artistic and like a professional shoot. Can’t wait to see what you come up with! Add some extra details after the trigger word for even better results. Scale I use: 1.2
If you’re stuck on the Kiro waiting list, you can just grab it right away using package managers. For Windows: winget install –id=Amazon.Kiro -e For macOS: brew install –cask kiro
🚀 Thank you for your amazing response to Kiro! We're seeing unprecedented demand. To ensure we can continue to provide the experience our community loves, we’ve implemented a waitlist as we fine tune Kiro. 👉 Sign up here spr.ly/60184htrA
Trying to get first-last frame alignment working in Seedance. Ended up with this random 2min POV Dog video while testing lol