Sasha Ivanov

super excited to share this “liquid photos” iOS app concept I’ve been working on! future user interfaces will feel a lot different than today created at @AXL_Labs

@earnestpost @1bajt 10/10 artisan human crafted lyrics

@mr_sudo_ @1bajt Hehe thank you! I wrote the lyrics myself >:)

Experimenting with music video generation again. I love the word meatbot so I fucked around on Suno and made this song, which then turned into this music video using the Grok video model mediated by @1bajt 's internal videogen software (that we're working on). Tell me your thoughts

@dodojin_xr @AXL_Labs thank you for the super kind words!!!

@mr_sudo_ @AXL_Labs 今天的网费没白花，看到如此酷的作品

instant pull-aparts! see exploded views of everyday objects with liquid photos. created at @AXL_Labs

@gabe__perez @AXL_Labs @ProductHunt we are definitely planning on it!

@mr_sudo_ @AXL_Labs Should def launch this on @ProductHunt once it’s open more generally

@notcnrad @AXL_Labs thank you!!

@mr_sudo_ @AXL_Labs beautiful work

@_themousepotato @AXL_Labs haha we were debating a couple of names, “pinch to explode” or “tear apart”

@mr_sudo_ @AXL_Labs So cool. Couldn't hold myself from recommending Dissect over Pull Apart for naming 😄

@KubaKotowicz @AXL_Labs you have a great point, the scene is only as good as the data we provide it, working on ways to get this context into the models before it generates

@mr_sudo_ @AXL_Labs Oh cmon… what kind of data? Repair manual with drawings? Schematics? 3d models or scans? Teardown videos?! That gamepad looks accurate-ish, but is the PCB layout real? If so, how does it know? If it’s not mostly accurate, then it’s party trick. Really cool UI/UX exploration tho

@hampusolsson @AXL_Labs The bottleneck is current gen ai video model speeds, in the tens of seconds at the moment for the quality we need for synthesizing transitions like this

@mr_sudo_ @AXL_Labs This is wild, does it do all the AI model generations on import? How long time does it take per image to process?

@WoubieSB @AXL_Labs We have people working, demo coming soon 👀

@mr_sudo_ @AXL_Labs Can we do it to the people or not yet?

here’s a link to the beta sign up for folks interested in trying it out! forms.gle/BgQfSqgeaAkbjF…

@ekherad_ @AXL_Labs beta is currently being developed, we’re working to get something out soon that people can have fun experiment with on their own photos!

@mr_sudo_ @AXL_Labs I signed up for the beta and didn’t receive any infor can you help me please?

@dankvr @AXL_Labs would go bonkers in webxr! i’m a mega @babylonjs fan myself 🥽

@mr_sudo_ @AXL_Labs This would go awesome with webxr 😁

@Must_be_Ash all pipelined up 👾

@mr_sudo_ Was it a one of or did you make a workflow/pipeline?

Can someone try this on their pet please? 🫪

@GadgetSignal @AXL_Labs give or take, everything aha

@mr_sudo_ @AXL_Labs I don’t even know what I’d try this on first. Probably every gadget in my house.

@KubaKotowicz @AXL_Labs DATA. the reality is the models need accurate references to compose these with higher precision

@mr_sudo_ @AXL_Labs How do you make it at least somewhat accurate?

@aretaidos @dennizor @AXL_Labs @TOtechweek really appreciate it thank you

@mr_sudo_ @dennizor @AXL_Labs @TOtechweek Absolutely, love your designs, truly inspiring

@BenThePearman @AXL_Labs hand model right here 💅

@mr_sudo_ @AXL_Labs Ay yo that's my 👆

@Must_be_Ash had fun making it, mostly seedance 2.0, for ones with human faces something like veo 3 is required

@mr_sudo_ Dude these transitions are smooth! What are the models being used for image and transitions video for the frame to frame?

@Aidan_Wolf @AXL_Labs I really appreciate this, yes! considering like a hyper zoom demo like this!! literally like zooming into a leaf like a bio textbook

@mr_sudo_ @AXL_Labs You're creating some great UX for how we may interact with objects in XR too. I can imagine 'zooming into' a plant with AR glasses and learning about photosynthesis

@Must_be_Ash Red Facts!

@mr_sudo_ i'm kinda scare to see Red in this :)