Positron AI lands $230 million in Series B funding, surpassing a $1 billion valuation to accelerate development of energy-efficient AI inference hardware.
Positron AI has secured an oversubscribed $230 million Series B funding round, pushing its post-money valuation past the $1 billion mark and placing the company firmly among the newest AI infrastructure unicorns. The round was co-led by ARENA Private Wealth, Jump Trading, and Unless, with strategic investments from the Qatar Investment Authority (QIA), Arm, and Helena. Existing backers, including Valor Equity Partners, Atreides Management, DFJ Growth, Resilience Reserve, Flume Ventures, and 1517, also participated, underscoring strong investor confidence in the company’s energy-efficient AI inference strategy.
The funding was announced at WebSummit Qatar and will be used to accelerate Positron’s product roadmap, moving beyond its current Atlas inference systems toward its next-generation custom silicon platform, Asimov. The company is targeting tapeout of the Asimov chip in late 2026, with volume production planned for early 2027.
Focus on energy-efficient AI inference
Positron is positioning itself as a next-generation infrastructure provider focused on lowering the cost and power consumption of AI inference at scale. According to CEO and cofounder Mitesh Agrawal, the market is increasingly constrained by power delivery and memory capacity rather than raw compute performance.
Agrawal said the company’s next-generation chips are designed to deliver up to five times more tokens per watt than NVIDIA’s upcoming Rubin GPU in key workloads. He also highlighted memory as a critical differentiator, noting that the Asimov silicon will feature more than 2,304 GB of RAM per device—significantly higher than the 384 GB expected in comparable systems. This expanded memory capacity is aimed at demanding applications such as video generation, algorithmic trading, multi-trillion-parameter models, and workloads requiring extremely long context windows.
Atlas systems already in market
Positron’s currently shipping product, Atlas, is an AI inference system built for rapid deployment and scaling. The system is manufactured in the United States, including both silicon and system-level production, allowing the company to ramp capacity quickly and provide supply chain reliability for customers.
Industry analyst Dilan Patel, founder and CEO of Semianalysis, noted that memory bandwidth and capacity are emerging as two of the biggest constraints in next-generation AI inference. He said Positron’s architecture is designed to address this challenge directly, with the Asimov chips expected to deliver more than ten times the high-speed memory capacity per chip compared with existing and emerging competitors.
Jump Trading’s dual role as customer and investor
A notable aspect of the Series B round is Jump Trading’s decision to co-lead the investment after first adopting Positron’s Atlas systems as a customer. According to Jump Trading CTO Alex Davis, the firm’s workloads are increasingly limited by memory and power rather than theoretical compute performance.
In internal testing, Davis said the Atlas platform delivered roughly one-third the end-to-end latency of comparable H100-based systems for the inference tasks evaluated. He also pointed to the system’s air-cooled design and production-ready footprint as advantages, along with a supply chain that simplifies procurement planning.
After reviewing Positron’s roadmap, including the Asimov silicon and the next-generation Titan system, Jump Trading chose to invest. Agrawal described the move as a strong endorsement, noting that a customer becoming a lead investor demonstrates both confidence in the technology and real-world demand.
Asimov and Titan: a memory-first platform
Positron’s upcoming Asimov chip is designed around the idea that modern AI workloads are increasingly constrained by memory bandwidth and capacity. Each accelerator is expected to support roughly 2 TB of memory, while the Titan system—built around the Asimov platform—will scale to about 8 TB per system. At rack scale, memory capacity could exceed 100 TB.
The company aims to match the effective memory bandwidth of NVIDIA’s next-generation Rubin GPUs while delivering better performance-per-watt. Arm, one of the strategic investors, emphasized the importance of system-level design and ecosystem collaboration in achieving these efficiency gains. The Asimov platform is built on Arm technology and is being developed in partnership with key ecosystem players, including Supermicro.
This memory-centric architecture is intended to support high-value inference workloads such as large-context language models, agent-based systems, and next-generation video and media models.
Fast development cycle
Positron says it is on track to tape out the Asimov chip just 16 months after securing the resources needed to start its design with Series A funding in June. The company intends to maintain this pace for future chip generations.
Agrawal emphasized that development speed is a critical competitive advantage in the AI chip race. To compete with industry leaders like NVIDIA, he said, Positron has structured its organization to deliver products on a similar cadence.
Growth outlook
Investors believe Positron is tackling one of the most pressing challenges in AI infrastructure: delivering inference at scale within real-world power and cost constraints. ARENA Private Wealth noted that the company’s early commercial traction with Atlas, combined with a clear path to Asimov, positions it to define a new category in AI hardware.
Positron expects strong revenue growth in 2026 and believes it could become one of the fastest-growing silicon companies in history. With multiple customers across cloud, advanced computing, and performance-intensive industries, the company is continuing to expand deployments and customer programs as it moves toward its next-generation platforms.
About Positron AI
Positron AI develops purpose-built hardware and software to dramatically lower the cost and improve the energy efficiency of AI inference. Positron’s currently shipping product, Atlas, is designed for rapid and scalable deployments. The company’s next-generation custom silicon, Asimov, is targeted for tapeout in late 2026 with production start in early 2027. Positron’s systems are designed to deliver superior economics for long-form contextual and next-generation AI workloads. Learn more at positron.ai .
The official version of this press release is the original language version. Translated versions are provided for the convenience of readers and have no legal effect. When using translated versions as reference material, please refer to the original language version, which is the only legally effective version.
