Positron AI, a leader in American-made semiconductors and inference hardware, announced the closing of a $51.6 million oversubscribed Series A round, raising over $75 million this year. Led by Valor Equity, Atreides Management, and DFJ Growth, with additional support from Flume Ventures, Scott McNealy, Resilience Reserve, 1517 Fund, and Unless. This funding supports the deployment of Atlas and the rollout of second-generation products in 2026.
With hundreds of billions expected to be spent on AI infrastructure in 2025, enterprises are facing rising costs, power limitations, and GPU shortages. Positron’s specialized solutions offer cost and efficiency benefits. Its Atlas delivers 3.5x performance per dollar and up to 66% lower power consumption than the NVIDIA H100, focusing solely on generative AI. Atlas’s FPGA architecture achieves 93% bandwidth utilization, supporting large models in a small server. It’s compatible with Hugging Face transformer models and accessible via an OpenAI API endpoint.
Powered by U.S.-fabricated chips, Atlas is in production, enabling LLM hosting and enterprise AI with lower latency and hardware needs. Founded in 2023 by CTO Thomas Sohmers and Chief Scientist Edward Kmett, with Mitesh Agrawal as CEO, Positron launched Atlas within 18 months, securing $12.5 million in seed funding and validating its performance, while gaining early customers. As adoption grows, Positron is developing custom ASICs to enhance performance, efficiency, and scalability. Its initial clients include Parasail, Cloudflare, and other major enterprises.
Following the Series A funding, the company is developing Titan, powered by ‘Asimov’ silicon, which supports up to 16 trillion parameters and large context sizes. Titan enables multiple models on a single system, with low-latency, high-throughput performance, making it easily fit into data centers.
KEY QUOTES:
“The early benefits of AI are coming at a very high cost – it is expensive and energy-intensive to train AI models and to deliver curated results, or inference, to end users. Improving the cost and energy efficiency of AI inference is where the greatest market opportunity lies, and this is where Positron is focused. By generating 3x more tokens per watt than existing GPUs, Positron multiplies the revenue potential of data centers. Positron’s innovative approach to AI inference chip and memory architecture removes existing bottlenecks on performance and democratizes access to the world’s information and knowledge.”
Randy Glein, co-founder and managing partner at DFJ Growth
“Memory bandwidth and capacity are two of the key limiters for scaling AI inference workloads for next-generation models. Positron is taking a unique approach to the memory scaling problem, and with its next-generation chip, can deliver more than an order of magnitude greater high-speed memory capacity per chip than incumbent or upstart silicon providers.”
Dylan Patel, founder and CEO of SemiAnalysis, and an advisor and investor in Positron
“We founded Positron to meet the demands of modern AI: aiming to run the frontier models at the lowest cost per token generation with the highest memory capacity of any chip available. Our highly optimized silicon and memory architecture allows for superintelligence to be run in a single system with our target of running up to 16-trillion-parameter models per system on models with tens of millions of tokens of context length or memory-intensive video generation models.”
Mitesh Agrawal, CEO of Positron AI
“We have passed on the overwhelming majority of AI accelerator startups we have diligenced over the last 6 years as most of them were mounting frontal assaults on NVIDIA that were unlikely to succeed. Positron has carefully chosen a defensible niche in low-cost inference. More importantly, they have proven that their software works before developing an ASIC: Positron is running competitively priced production inference workloads today on 2022 era FPGAs in a server of Positron’s own design. This speaks to the quality of their software stack, their system-level expertise and the judgment of their management team.”
Gavin Baker, managing partner and chief investment officer of Atreides Management
