Positron AI Raises $51.6 mn to Rethink Inference Hardware for the AI Era

AI hardware startup Positron AI has raised $51.6 million in an oversubscribed Series A funding round, bringing its total funding this year to over $75 million.

The funding will accelerate the rollout of its debut product, Atlas, and support development of its next-generation system, Titan, slated for 2026.

As AI infrastructure costs soar and NVIDIA GPU supply tightens, Positron is positioning Atlas as a cost-efficient alternative for generative AI workloads. The company claims Atlas offers 3.5x better performance per dollar and 66% lower power usage than NVIDIA’s H100.

“Our highly optimised 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, said.

Atlas uses a memory-optimized FPGA architecture that supports models with up to 500 billion parameters and delivers 93% memory bandwidth utilization—far above the 10–30% of typical GPUs. It runs LLMs through an OpenAI-compatible API and is already being used for hosting, copilots, and generative agents.

The round was led by Valor Equity Partners, Atreides Management, and DFJ Growth, with participation from Flume Ventures, Resilience Reserve, 1517 Fund, and Unless.

Positron was co-founded in 2023 by CTO Thomas Sohmers and Chief Scientist Edward Kmett, with former Lambda COO Mitesh Agrawal joining as CEO to scale the company’s commercial operations.

In just 18 months, the team brought Atlas to market with only $12.5 million in seed funding. They validated performance, landed early enterprise customers and hardened the product in deployment environments before raising this Series A.