Developing the next generation of machine learning hardware and software
Reno, NV
Joined August 2023
- Tweets94
- Following56
- Followers2.4K
- Likes48
Thank you, @swyx, for the thoughtful words about our Co-Founder and CTO @trsohmers on @AI_in_the_AM. We appreciate you taking the time to share them—and the conversation.
Positron AI retweeted
Thomas Sohmers, whose servers run at Oracle, on a planned chip with eight GPUs' worth of memory:
"With Asimov, our upcoming chip, we have up to 2.3 terabytes of memory capacity per chip."
He compares that with the B300's 288 gigabytes. The eight-GPU comparison is about memory capacity.
"We have to use the commodity memory because that's the only thing that's going to be able to scale and be cost effective."
He puts NVIDIA GPUs at roughly 30–40% memory bandwidth utilization when generating text.
"With our first gen product, we were able to hit and sustain 93% of theoretical memory bandwidth."
"Whenever you have to scale to having more than one device, there's overheads involved with that."
Keeping the model on one device removes some software and deployment complexity, as well as performance overhead—not just the cost of additional chips.
@trsohmers
Larger models, longer context, more agents working in parallel. Memory needs to scale with all three.
Our Co-Founder and CTO @trsohmers explains how those demands shaped Positron’s approach to commodity memory.
Watch the @AI_in_the_AM clip.
🎤 Happening now at #Yotta2026 in Las Vegas: our CEO @mitesh711 takes the stage for “The Token Wars: Founders Building the Inference Era.” The conversation: what comes after the training boom—and how inference is reshaping the future of AI compute. #AI #AIInfrastructure #Inference #FutureOfCompute #PositronAI
Our CEO @mitesh711 joined @sophiadew on @MTSlive to discuss our $875M Series C, growing AI inference demand, and how Positron is scaling more efficient infrastructure.
Thanks to Sophia and the MTS team!
Watch Mitesh’s interview at 5:51:39: youtube.com/watch?v=GUhMwp32…
Scaling AI means making every dollar—and every watt—count.
Our own @chasiu_ joined @furrier on @theCUBE + @NYSEWired to discuss tokens per dollar, tokens per watt, and matching inference hardware to the workload.
Watch the interview: youtu.be/_keLh5gFEM0?si=plW_…
What would you build if AI were dramatically cheaper to run?
Our Co-Founder @trsohmers joined @HarryStebbings on @20vcFund to explore the hardware, memory and economics behind AI inference—and what comes next.
Watch the episode: thetwentyminutevc.com/thomas…
Yesterday at the AI Infrastructure Summit, our CEO @mitesh711 shared what it takes to build a semiconductor company for the inference era. Special guest Jay Jackson of @OracleCloud joined him to discuss the infrastructure challenges of growth Positron’s work to bring its systems online alongside GPUs at Oracle Cloud.
Today, our CEO @mitesh711 chatted with @jordihays and @johncoogan live on @tbpn to discuss our $875 million Series C.
youtube.com/live/0VEqt0pwpFI… via @YouTube
The conversation covered Positron’s journey, the rapidly growing demand for AI inference and how we’re scaling to meet what comes next.
Thanks to Jordi, John and the entire team for a great conversation.
Positron AI retweeted
Thrilled to announce @positron_ai's $875M Series C at a $5B valuation!
This round was co-led by @NEA, @Atreidesmgmt (@GavinSBaker), @valor (@AntonioGracias), @AndraCapital, @SemiAnalysis_ Capital (@dylan522p), and Jim Clark (Founder of SGI and Netscape). An incredible group to have backing what we’re building.
Also very happy to have strategic backing from VentureTech Alliance, Hudson River Trading, Cisco Investments, and Naver Ventures, alongside continued support from @Arm , @jumptrading , @dfjgrowth, Qatar Investment Authority, and many more.
What I’m most excited about is what this lets us do next. We’re deploying more than 50 racks of Atlas at @OracleCloud, and this financing funds our next-generation Asimov chip’s tapeout, the production ramp for Titan, and the manufacturing capacity to meet the demand ahead of us.
The goal remains simple: make running AI dramatically cheaper and more energy efficient.
A huge thank you to the entire Positron team, especially @mitesh711 and @kmett, our customers, partners, and investors. Building hardware takes an enormous amount of work, and I couldn’t be more proud of this team. We have a lot ahead of us, and I can’t wait to share more.
The inference era is here. Today, we announced Positron AI’s $875M Series C at a $5B valuation.
Our CEO, @mitesh711, spoke with @RWhelanWSJ about why inference is reshaping AI infrastructure, our momentum and what we’re building next.
Read more: wsj.com/tech/ai/positron-val…
Our CEO, @mitesh711, spoked with Rob Sabo at NNBW about where Positron is heading in 2026.
The point we keep coming back to: in this industry, product cadence matters as much as product quality. What customers teach us on Atlas is already shaping Titan, and that loop is how a team of under 50 competes with trillion dollar incumbents.
Shipping today, going live with a hyperscaler next, and hiring fast to keep up. Full article in the comments. More to come.
Positron AI retweeted
NVIDIA has a monopoly. Thomas Sohmers is unbothered.
@JordanNanos sits down with the Co-Founder & CTO of @Positron_AI to talk FPGA inference, LPDDR memory, and running 16T parameter models on a single box.
@trsohmers
Tune in: youtu.be/B8O3pLcX2w4
Positron AI retweeted
Most people think AI scaling is just about more compute. They’re wrong. ❌
The real bottleneck? Memory.
In this exclusive clip from GTC 2026, Thomas Sohmers (CEO of @positron_ai) breaks down how their new "Asimov" chip is flipping the script:
- 16 Trillion parameter models running on a single server.
- Millions of tokens in context length.
- Massive realized memory bandwidth that reduces the need for complex networking.
If you want to know what the next phase of LLM infrastructure actually looks like, you need to watch this.
👇 Full conversation in the replies!
Check out the interview with our CEO @mitesh711 by @ndahad from @eetimes at the Arm AGI CPU Launch Event: youtube.com/watch?v=3yOnilgE…