MatX secures $500 million series B to advance AI training chips

AI hardware startup MatX has raised $500 million in a Series B funding round, underscoring growing investor interest in specialized processors for artificial intelligence model training. The round was led by Jane Street, with participation from Spark Capital, Marvell Technology, the Stripe co founders, and several other strategic investors. The capital will support the development, testing, and scaling of MatX’s high-performance AI training chips, with initial shipments expected in 2027. The move signals confidence in MatX’s ability to compete in a market dominated by traditional GPU providers.

MatX series B funding and AI hardware ambitions

Founded in 2023 by former Google engineers Reiner Pope and Mike Gunter, MatX focuses on building processors purpose-built for training large-scale AI models. Their flagship product, the MatX One chip, is designed to deliver high throughput and energy efficiency, addressing the growing computational demands of advanced neural networks and large language models.

Investors were attracted to MatX’s deep technical expertise and market potential. Jane Street took the lead in the round, providing both capital and strategic oversight, while Spark Capital and other participants contribute experience in scaling high-tech startups. The funding will accelerate final chip design, prototyping, and large-scale manufacturing through a partnership with TSMC, the world’s largest contract semiconductor foundry. MatX anticipates that production-ready chips will be available by 2027 to meet increasing enterprise and research demand.

In addition to manufacturing, the capital will also support software development and optimization, ensuring MatX chips integrate seamlessly with AI frameworks and workloads. This software-hardware synergy is critical for enterprises that need flexible solutions to train complex AI models efficiently.

Market context and AI hardware trends

The Series B funding reflects broader trends in the AI industry, where demand for specialized AI hardware is rising sharply. Large-scale AI models, particularly those used in natural language processing, recommendation systems, and generative AI, require processing power far beyond conventional GPUs. Purpose-built chips like MatX’s promise higher efficiency, lower energy consumption, and faster training times.

While NVIDIA and other GPU manufacturers dominate the AI training market, startups like MatX aim to carve out a niche by optimizing performance specifically for large-model workloads. Analysts note that AI hardware startups are increasingly attractive to investors, given the projected growth of AI applications across industries such as healthcare, finance, robotics, and autonomous systems.

The funding also signals that institutional and strategic investors are willing to place significant bets on AI hardware as a critical component of AI infrastructure, not just software innovation. By providing the capital to bring next-generation processors to market, investors are positioning themselves to benefit from both short-term adoption and long-term industry transformation.

MatX Background and Vision

MatX was founded to combine expertise in hardware architecture and AI software, leveraging insights gained from Google’s Tensor Processing Unit (TPU) program. CEO Reiner Pope and CTO Mike Gunter bring extensive experience in designing AI accelerators optimized for large-scale workloads. MatX’s architecture uses advanced memory hierarchies, parallel processing arrays, and energy-efficient design principles to achieve high performance while controlling power consumption.

The company emphasizes flexibility and software compatibility, allowing its chips to support a broad range of AI workloads. MatX is positioning itself not just as a chipmaker but as a partner for enterprises and AI researchers seeking scalable, high-efficiency hardware solutions. By combining cutting-edge design with strategic investor support, the startup aims to deliver a compelling alternative to existing GPU-based training platforms.

MatX’s $500 million Series B funding highlights the growing importance of AI hardware startups in the broader AI ecosystem. With backing from major investors, advanced chip designs, and manufacturing partnerships, the company is set to compete in the high-demand market for AI training processors. As large-scale AI models continue to proliferate, MatX’s progress will be closely watched by investors, enterprises, and competitors alike, potentially shaping the next generation of AI compute infrastructure.