Infinity AI
@toinfinityai
YC W24. A script-to-movie foundation model. Our first product allows people to generate talking-head videos. Try it here: studio.infinity.ai
ID: 1453384705349410827
http://studio.infinity.ai 27-10-2021 15:34:59
244 Tweet
3,3K Takipçi
47 Takip Edilen
Congrats to user Sojin Samuel for winning 2nd place in this week's video competition! 🏆 Check out this zombie summer news segment 🧟♀️☀️
“Azure has the cheapest GPUs compared to any of the other major providers.” “InfiniBand is incredibly easy to set up.” A big thank you to Lina Colucci for this insightful video on best practices for training large transformer models and scaling up your training efforts. We
Congrats to #Crypto #Cam for the honorable mention in the competition this week! 👏
This took me right back to Y Combinator office hours with Michael Seibel 🤣 Garry Tan’s roast is spot on too. Great job, Lina Colucci and Infinity AI 🙌 Check it out: twitter.infinity.ai
The folks at Infinity AI sponsor a weekly video contest using their video generator. I won the last 2 weeks, so here is my entry for this week's contest. The theme is "Roasts" which can be interpreted multiple ways, but you know where I'm going. Fingers crossed for win #3.
the day when the default ElevenLabs audio for joey ("an upbeat news host") makes a lot of sense 😂
Diffusion transformers are awesome, but while we all wait for Sora, I like Infinity AI's approach - severely constrain the usecase to just video sync (not just lip sync) - and go from there. Because of subtle head (and even jewelry) movements, this thing immediately leaps
We posted on Hacker News today! Here are some cool videos we've seen our model generate so far 🧵 1\ The Sora lady speaks!
The beauty in AI art is when it makes mistakes. Here's "Marilyn & John" Workflow: - Libsync by Infinity AI - Lyrics & Prompts by ChatGPT - Music by Udio ➡️ VocalRemover ➡️ Remaster in Ableton - Images by Flux - Video by RunwayML - Resolve: PlanarTracker, super scale 2X