California’s OXMIQ raises $35M to expand AI chip architecture

California-based OXMIQ has raised $35 million in Series A funding to accelerate development of its licensable AI chip architecture, bringing the company’s total funding to $60 million. The Campbell-based startup, founded by semiconductor veteran Raja Koduri, is developing GPU and AI technologies designed to help semiconductor companies and system builders create custom AI chips more quickly and at lower cost. The funding round was co-led by Fundomo and Samsung Catalyst Fund, with participation from MediaTek, AM Intelligence Labs, Pegatron Venture Capital, CDIB-TEN, Darwin Ventures, Morgan Creek Digital, and other strategic investors. OXMIQ said the investment will be used to scale OxCore™, its GPU architecture that enables chipmakers to build customised AI silicon without developing an entirely new chip from scratch. The company is targeting one of the fastest-growing segments of the AI industry as demand for AI computing infrastructure continues to outpace semiconductor supply. At the centre of OXMIQ’s platform is OxCore™, a licensable GPU architecture that combines graphics processing, AI tensor computing, and CPU orchestration into a single scalable platform. The company says the approach reduces development costs while improving performance and energy efficiency for AI workloads. OXMIQ also offers OxQuilt™, a chiplet integration platform that allows customers to combine processors, memory, and packaging technologies from multiple suppliers, providing greater flexibility than traditional chip designs. Supporting the hardware is a software platform that enables developers to run existing CUDA and PyTorch applications without modifying code, helping organisations integrate AI workloads across different computing environments. The company is pursuing an …

California-based OXMIQ has raised $35 million in Series A funding to accelerate development of its licensable AI chip architecture, bringing the company’s total funding to $60 million.

The Campbell-based startup, founded by semiconductor veteran Raja Koduri, is developing GPU and AI technologies designed to help semiconductor companies and system builders create custom AI chips more quickly and at lower cost.

The funding round was co-led by Fundomo and Samsung Catalyst Fund, with participation from MediaTek, AM Intelligence Labs, Pegatron Venture Capital, CDIB-TEN, Darwin Ventures, Morgan Creek Digital, and other strategic investors.

OXMIQ said the investment will be used to scale OxCore™, its GPU architecture that enables chipmakers to build customised AI silicon without developing an entirely new chip from scratch.

The company is targeting one of the fastest-growing segments of the AI industry as demand for AI computing infrastructure continues to outpace semiconductor supply.

At the centre of OXMIQ’s platform is OxCore™, a licensable GPU architecture that combines graphics processing, AI tensor computing, and CPU orchestration into a single scalable platform. The company says the approach reduces development costs while improving performance and energy efficiency for AI workloads.

OXMIQ also offers OxQuilt™, a chiplet integration platform that allows customers to combine processors, memory, and packaging technologies from multiple suppliers, providing greater flexibility than traditional chip designs.

Supporting the hardware is a software platform that enables developers to run existing CUDA and PyTorch applications without modifying code, helping organisations integrate AI workloads across different computing environments.

The company is pursuing an intellectual property licensing model rather than manufacturing complete chips, allowing semiconductor companies to incorporate OXMIQ’s technology into their own products while reducing development time and capital requirements.

David Goldschmidt, Managing Director of Samsung Catalyst Fund, said OXMIQ’s architecture addresses growing demand for more efficient AI infrastructure.

“OXMIQ’s novel AI core and software platform enable heterogeneous compute for efficient, custom inference solutions serving large-scale AI workloads,” he said.

Founder and CEO Raja Koduri said lowering the cost of AI computing will make advanced AI technologies more widely accessible.

“A licensable core with an open architecture means design teams everywhere can build the custom AI silicon their work needs,” Koduri said.

Alongside the funding announcement, OXMIQ strengthened its leadership team with the appointment of Tenstorrent CEO Jim Keller to its board of directors. Widely regarded as one of the semiconductor industry’s leading chip architects, Keller joins the company as it moves from platform development to customer deployment.

The latest investment reflects growing venture capital interest in next-generation AI infrastructure as demand for specialised semiconductors continues to accelerate. Rather than competing directly with established chip manufacturers, OXMIQ is positioning itself as a technology provider that enables a broader range of companies to develop custom AI hardware for emerging applications.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

Keep in touch with our news & offers

Subscribe to Our Newsletter

Previous Post California-based startup Hakimo raises $12M to expand AI security
Next Post INDONESIA 2010

Enjoy Unlimited Digital Access

Read trusted, award-winning journalism. Just $2 for 6 months.
Already a subscriber?
Share the post