Positron Secures $23.5M to Expand U.S.-Made AI Chip Production

Feb 12, 2025
Positron has raised $23.5 million to scale production of its energy-efficient AI chips, offering a cost-effective alternative to Nvidia's hardware.
Positron Secures $23.5M to Expand U.S.-Made AI Chip Production

Positron, a company specializing in AI inference technology, has secured $23.5 million in funding from investors including Flume Ventures, Valor Equity Partners, Atreides Management, and Resilience Reserve. This funding will support the expansion of Positron's production of energy-efficient AI chips, which are manufactured entirely in the United States.

Positron's AI chips are designed to provide a cost-effective alternative to Nvidia's hardware, achieving 3.5 times better performance per dollar and greater power efficiency. The company's Atlas systems utilize a memory-optimized architecture, supporting trillion-parameter models and offering compatibility with Hugging Face and OpenAI APIs. These systems deliver faster inference and lower power consumption compared to Nvidia's H100 GPUs.

The company is led by CEO Mitesh Agrawal, who joins co-founders Thomas Sohmers and Edward Kmett. Positron's focus on domestic manufacturing aims to reduce reliance on foreign supply chains, positioning the company as a key player in the U.S. AI hardware market. The funding will also aid in the development of Positron's second-generation accelerators, expected to launch in 2026.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Silicon Brief, AI Funding Brief or Daily AI Brief.

Free newsletter

Silicon Brief

Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more
Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.