World Models @ Reactor / ex-Adept AI / ex Amazon AGI Lab ⛰️ Alpine enthusiast
San Francisco, CA
Joined December 2025
- Tweets68
- Following173
- Followers55
- Likes381
Kavya retweeted
Join us next week at Long Horizon Summit by @encord_team and learn about robotics on Reactor!
📅 Oct 20–21
📍 Fort Mason, San Francisco
🤖 Booth SU #1
There may or may not be limited edition Reactor swag if you track us down...
Kavya retweeted
Introducing FastH3-Omni.
The fastest and highest-quality open-source version of realtime H3 by @MiniMax_AI, post-trained by FastVideo @haoailab with @nuvalab.
Exclusive preview on the Reactor API until 10.20.2026.
• First and last frame conditioning
• Audio input + strong dialogue
• Up to 3 reference images
• Frontier-level performance
FastH3 is now live on @reactorworld!! Check it out!
Introducing MiniMax FastH3 on Reactor - the latest open-source model by @MiniMax_AI.
Stream 720p with audio at $0.007/second with first frame image support, faster and cheaper than ever.
Get started now by cloning our demo app: reactor.inc/fast-h3-demo-app
Really curious about one-shot learning, especially how far it generalizes across different environments and embodiments as physical pretraining scales.
World Action Models aren’t just about what to predict, it’s also about how fast you can act on that prediction. How does latency change the equation? Come chat with me at #Actuate!
Reactor for Robotics will be at Fort Mason for Actuate!
Come and meet our team if you're working at the intersection of robotics and world models 🤖
actuate.foxglove.dev/
Kavya retweeted
Introducing LTX’s newest open-weight model for realtime avatars - available on Reactor day 0.
We're proud to partner with @ltx_io to make it easier than ever to build lip-synced, fully-voiced AI personas at cinematic quality with our API.
Try it here: reactor.inc/ltx
Kavya retweeted
Introducing OpenDreamer, an open reproduction of Dreamer4.
We supported @FrancescoSacco1, @martidiegom, and @edward_s_hu to build a frontier-level world model, and we've open-sourced everything: the model, the training code, and every hard-won lesson from building it.
Happy Oyster live on @reactorworld! Try it out now!
🚀 Introducing Happy Oyster, the groundbreaking real-time world model from @HappyOysterAI, live today on Reactor!
Explore any world you can imagine, or take the director's chair and shape the story as it happens.
Available via API.
Try it now: reactor.inc/happy-oyster
Kavya retweeted
🚀 Introducing Lingbot-World-2, our newest real-time world model.
Richer interactivity, sharper visuals, and advanced camera controls that make exploring what you create feel truly alive.
Available today via API, exclusively on Reactor.
Try it now: reactor.inc/lingbot-world-2
Kavya retweeted
Real-time video editing is here.
SANA-Streaming by @NVIDIA is now live on Reactor. Change the style, the scene, the mood of any video as it plays, with no rendering wait.
Try it now: sana-streaming.reactor.inc
Kavya retweeted
Organizing the top World Models summit in the bay this month: luma.com/hkik7xtr
be there or be square
Physical AI is having its moment. Robotics, autonomous vehicles, interactive video. Underneath all of it sits the same primitive: a world model that predicts how an environment responds to action.
There's a lot of debate about world models. JEPA vs. diffusion vs. autoregressive. How sim-to-real actually transfers. Nobody agrees on how to evaluate physical understanding.
So @taiuti & @_bschmidtchen at @reactorworld , @ashray_malhotra at @Adobe, and I decided to put the top minds in the field in one room. On June 18 in SF, we're hosting World Models SF Summit, likely the first summit dedicated entirely to world models.
Researchers, founders, and practitioners from @GoogleDeepMind @OpenAI @Meta @nvidia @theworldlabs @Tesla @Apple @Microsoft @Roblox and many strong startups in the space have already signed up.
A few seats left: luma.com/hkik7xtr
So great to collaborate with the researchers at @nvidia to help serve this model! 🚀
🚀 Causal realtime streaming SANA-WM open-sourced!
Thanks to @reactorworld for serving the model — try demo: sana-wm.reactor.inc/
~0.93x realtime on single H100, watch 60s 720p live + 6-DoF camera control.
Code: nvlabs.github.io/Sana/docs/s…
Kavya retweeted
Time to 𝗿𝗲-𝗺𝗲𝗲𝘁 𝗟𝗶𝗻𝗴𝗯𝗼𝘁-𝗪𝗼𝗿𝗹𝗱-fast — powered by @reactorworld 🌍!
Lingbot is one of the most powerful open-source world models in the community, from the Robbyant Team.
We got it running in real-time — and, just as importantly, made it genuinely fun and easy to use.
🔗 github.com/Robbyant/lingbot-…
Kavya retweeted
Today Reactor is coming out of stealth. We’ve raised $59M in Seed and Series A funding, led by @lightspeedvp, with participation from @AmplifyPartners, @wndrco, @Sky9Capital, and @FPVventures.
Reactor is the platform for building in the World Model era: the infrastructure that lets developers build with them at global scale for the first time. Stream from a frontier World Model to your app, in real time, all in under 10 lines of code.
World Models represent the next major shift in AI: pixels, audio and actions are generated on the fly, in real-time, in response to user inputs, and to the environment. Every time computing has made a shift from passive to interactive, entire industries appeared that didn't exist before. We're standing in front of such moment again.
Over the last 6 months, we’ve assembled an all-star team with alumni from Apple, Meta, Google, Luma AI, Netflix, and Replicate. We're already partnering with some of the biggest names and labs in the world, and hundreds of developers are already building on Reactor.
The World Model era starts now.
Kavya retweeted
It’s time to unveil what we’ve been working on: today, @reactorworld comes out of stealth with $59M to power the World Model era.
And we’re just getting started.
Today Reactor is coming out of stealth. We’ve raised $59M in Seed and Series A funding, led by @lightspeedvp, with participation from @AmplifyPartners, @wndrco, @Sky9Capital, and @FPVventures.
Reactor is the platform for building in the World Model era: the infrastructure that lets developers build with them at global scale for the first time. Stream from a frontier World Model to your app, in real time, all in under 10 lines of code.
World Models represent the next major shift in AI: pixels, audio and actions are generated on the fly, in real-time, in response to user inputs, and to the environment. Every time computing has made a shift from passive to interactive, entire industries appeared that didn't exist before. We're standing in front of such moment again.
Over the last 6 months, we’ve assembled an all-star team with alumni from Apple, Meta, Google, Luma AI, Netflix, and Replicate. We're already partnering with some of the biggest names and labs in the world, and hundreds of developers are already building on Reactor.
The World Model era starts now.
Kavya retweeted
I realized that what I cannot profile, I cannot optimize.
This is why I embarked on a little project in Diffusers, to try to profile important pipelines, identify bottlenecks for torch.compile, and fix them. Got decent results.
I documented the process and invited the community to apply the same.
@ariG23498 decided to take it a notch further by formulating an entire series of tutorials around the topic, starting from compiling simple torch ops and how to make sense of their profile traces.
Follow his space to stay updated.
It's an incredibly helpful skill to have, especially if you're in the optimization business. Even if you're not, it gives a good mental model of what's going on in those SMs.
Kavya retweeted
New blackboard lecture w @reinerpope
How do chips actually work – starting with basic logic gates, and working up to why GPUs, TPUs, FPGAs, and the human brain each look the way they do.
0:00:00 – Building a multiply-accumulate from logic gates
0:16:20 – Muxes and the cost of data movement
0:25:59 – How systolic arrays work
0:39:00 – Clock cycles and pipeline registers
0:51:40 – FPGAs vs ASICs
1:03:14 – Cache vs scratchpad
1:07:16 – Why CPU cores are much bigger than GPU cores
1:11:49 – Brains vs chips
1:15:22 – A GPU is just a bunch of tiny TPUs
Look up Dwarkesh Podcast on YouTube/Spotify/etc to watch. Enjoy!
Kavya retweeted
The future is arriving faster than you think
Took 2 of my nephew's drawings and made them into worlds he can walk inside.
Now he says, "Make it rain," and the sky cracks open.
1 hour at the airport, with @reactorworld
This is what AI is actually for.