Reno’s Positron Joins the AI Chip Unicorn Club

Reno startup Positron AI reaches unicorn status with a $230M round to scale its US-built chiplet architecture for AI inference

9 Mar 2026

Positron AI, a Nevada-based semiconductor startup, has raised $230mn in a Series B funding round, valuing the company at more than $1bn as investors back new approaches to artificial intelligence inference hardware.

The Reno-based group develops AI chips designed primarily for inference, the process of running trained models in production. The round was backed by Arm, Qatar Investment Authority and trading firm Jump Trading, which joined as a co-lead investor after previously using the company’s hardware as a customer.

The funding comes as demand for inference computing grows across cloud providers, financial institutions and enterprise software groups. Companies deploying large language models are increasingly looking for alternatives to graphics processing units supplied by Nvidia, which dominate the market.

Positron’s current product, the Atlas chip, is already being manufactured in Arizona and shipped to customers. The company says systems built around Atlas deliver roughly three times lower latency on inference workloads compared with configurations using Nvidia’s H100 chips, while using less power.

The company’s next processor, known as Asimov, is designed around a chiplet architecture, a method of assembling processors from multiple smaller silicon components rather than a single monolithic die.

Positron says Asimov will integrate 20 memory “fanout gearbox” chiplets and support more than 2,300GB of RAM per accelerator. Greater memory capacity has become a key constraint in running large AI models efficiently, particularly in enterprise deployments.

Tape-out, the stage when the chip design is finalised for manufacturing, is scheduled for late 2026, with production targeted for early 2027.

The funding also reflects growing interest from sovereign investors in securing access to advanced computing infrastructure. Qatar Investment Authority’s participation highlights efforts by state-backed funds to support emerging semiconductor supply chains beyond traditional manufacturing hubs.

For the US semiconductor sector, the investment signals increasing investor confidence in startups pursuing chiplet-based designs as a way to compete in the rapidly expanding market for AI inference hardware.