Thinking Machines Lab releases preview of real-time interaction models
Thinking Machines Lab announced a research preview of interaction models built from scratch for continuous real-time collaboration. The models process simultaneous audio, video, and text streams through a multi-stream micro-turn architecture. Founder Mira Murati noted that current AI systems force users to adapt to turn-based limitations rather than enabling fluid exchanges. A demonstration video shows the system offering timely audio suggestions during live conversations.
The current “AI experience” often feels like a conversation that only begins after we stop talking. We have to batch our thoughts. We can’t point at things. We phrase questions like emails. The interface doesn't leave room for us so we adapt to the models.
Today we're sharing our work on interaction models. A new class of model trained from scratch to handle real-time interaction natively, instead of gluing it onto a turn-based one. https://youtu.be/A12AVongNN4
Today we're sharing our work on interaction models. A new class of model trained from scratch to handle real-time interaction natively, instead of gluing it onto a turn-based one. https://youtu.be/A12AVongNN4
We started Thinking Machines to advance human-AI collaboration, and this is our first bet on what that looks like. Most labs treat autonomy as the goal and interactivity as scaffolding around a turn-based core. We think the way we work with AI matters as much as how smart it is. Interactivity has to be in the model, and it has to scale with intelligence rather than trail behind it. https://thinkingmachines.ai/blog/interaction-models/
The current “AI experience” often feels like a conversation that only begins after we stop talking. We have to batch our thoughts. We can’t point at things. We phrase questions like emails. The interface doesn't leave room for us so we adapt to the models.
I believe the kids call this "@thinkymachines just brutally framemogged gdm and oai".
basically everyone's definition of "realtime" just got a massive frciking upgrade
lowkey the funniest videos of the batch. thinky has some comedians!! congrats to @thinkymachines on reviving the omnimodel dream that others could not
In modern ML accelerators, FLOPS have absolutely exploded. Often though, the bottleneck is not FLOPS but memory bandwidth. Similarly, model intelligence has exploded, causing the bottleneck to be human<->AI bandwidth. At Thinky, we think that it’s important to solve this. 1/4
my highlights
lowkey the funniest videos of the batch. thinky has some comedians!!
congrats to @thinkymachines on reviving the omnimodel dream that others could not
Today we're sharing our work on interaction models. A new class of model trained from scratch to handle real-time interaction natively, instead of gluing it onto a turn-based one. https://youtu.be/A12AVongNN4
@thinkymachines her. this is her.
lowkey the funniest videos of the batch. thinky has some comedians!! congrats to @thinkymachines on reviving the omnimodel dream that others could not
@miramurati That is awesome.
Today we're sharing our work on interaction models. A new class of model trained from scratch to handle real-time interaction natively, instead of gluing it onto a turn-based one. https://youtu.be/A12AVongNN4
The demo of this model is cool.
Interacts with multiple people in multiple languages. On multiple tasks.
Now we know why @miramurati got the big bucks.
Today we're sharing our work on interaction models. A new class of model trained from scratch to handle real-time interaction natively, instead of gluing it onto a turn-based one. https://youtu.be/A12AVongNN4
@miramurati Congratulations @miramurati! This looks amazing 🤩
Today we're sharing our work on interaction models. A new class of model trained from scratch to handle real-time interaction natively, instead of gluing it onto a turn-based one. https://youtu.be/A12AVongNN4
@miramurati Congratulations! Looks quite cool!
Today we're sharing our work on interaction models. A new class of model trained from scratch to handle real-time interaction natively, instead of gluing it onto a turn-based one. https://youtu.be/A12AVongNN4
@miramurati Very proud of your hard work Mira. All of our sacrifice has been worth it.
Today we're sharing our work on interaction models. A new class of model trained from scratch to handle real-time interaction natively, instead of gluing it onto a turn-based one. https://youtu.be/A12AVongNN4