Colin Delea

GP at Flume Ventures | Connecting visionary founders with our community of operators

[Positron AI @positron_ai]

Why we built Positron.

Simple: To enable Superintelligence for everyone through the best performance-per-dollar and per-watt systems for generative AI inference.

Positron’s first-generation product, Atlas, is shipping now, and we are happy to announce that Positron’s second-generation product will ship in 2026.

Positron is focused on maximizing memory bandwidth and memory capacity on our Terachip (up to 2TB memory per chip). See what we have built and what we are building:

[Mitesh Agrawal @mitesh711]

We are excited to announce that @positron_ai has raised $51.6M Series A led by Valor Equity Partners, @Atreidesmgmt and @dfjgrowth with participation from Resilience Reserve, Flume Ventures, 1517 Fund and Unless. We are looking forward to working closely with @GavinSBaker , @rglein and @vpat24 to deliver the best performance/$ and performance/watt system for frontier models. More details on what we are up to below:

[Thomas Sohmers @trsohmers]

My startup, @positron_ai just raised $51.6M Series A to rebuild the infrastructure powering AI inference and bring Superintelligence to everyone.

Led by Valor Equity Partners (@valorep) , Atreides Management (@Atreidesmgmt) , and DFJ Growth (@dfjgrowth) —the same teams behind SpaceX, Tesla, xAI, and other companies pushing compute to its physical and economic limits.

Why inference? Because as AI moves from research into production, the real bottleneck isn’t training—it’s deploying transformer models at scale. GPUs were built for flexibility, not predictable workloads. That mismatch means wasted bandwidth, poor density, and massive power costs.

So at Positron, we built silicon specifically optimized for inference:
•>90% memory bandwidth utilization (GPUs typically ~30%)
•Multi-model hosting per card (higher density, lower power)
•Zero code changes — drop in compatability with the HuggingFace Transformers ecosystem.

U.S.-made silicon, stable supply chain, geopolitically resilient
We’re already shipping and running in production environments today.

Here’s what we’ve built, and link to our WSJ coverage and press release in the follow up reply ↓