Infinity Secures $15M Seed to Make Any AI Chip Instant-Inference Ready

SAN FRANCISCO — July 21, 2026 — Infinity.inc (Infinity Artificial Intelligence Institute) today announced a $15 million Seed funding round at a $100 million post-money valuation. The round was led by Touring Capital, with participation from Principal VC, executives at major chip companies, and AI researchers from OpenAI and Anthropic.

The capital will scale Infinity’s engineering team and accelerate the deployment of its automated AI research platform and core engine, Ignition.

Breaking the CUDA Lock-in Bottleneck

NVIDIA’s dominance in AI compute relies heavily on its proprietary CUDA software stack. While specialized semiconductor companies produce competitive accelerators, GPUs, and custom silicon, adopting them traditionally requires months or years of manual engineering to write custom low-level software kernels.

Infinity eliminates this software friction with Ignition, an autonomous AI research agent designed to write, test, debug, and optimize low-level inference code across any hardware architecture.

“The AI industry has operated under an artificial constraint that only a handful of chips could run AI well, because only NVIDIA spent decades building the software to make them work,” said Jeremy Nixon, Founder and CEO of Infinity. “Ignition eliminates that constraint. The next era of AI will be defined by who can make any chip run state-of-the-art models at blazing speeds.”

How Ignition Transforms Hardware Deployment

By automating low-level compiler and kernel engineering, Infinity makes alternative silicon inference-ready in days:

• Automated Kernel Generation: Ignition writes tailored inference software for diverse chip architectures, including GPUs, SRAM, mobile processors, and systolic arrays.
  
  
• Self-Optimizing Benchmarks: The agent continuously evaluates throughput (tokens per second), debugging and rewriting code iteratively until performance targets are met.
  
  
• Performance-Aligned Business Model: Infinity ties revenue directly to customer outcomes, taking a share of measurable speed improvements and cloud cost savings.

Commercial Momentum and Industry Support

Infinity is already generating millions of dollars in Annual Recurring Revenue (ARR) through design partnerships with chip manufacturers, including d-Matrix.

“Software, not raw silicon design, is the primary barrier holding back alternative hardware adoption in AI compute,” said investors at Touring Capital. “Infinity’s automated execution layer unlocks the true performance potential of the broader semiconductor ecosystem.”

About Infinity.inc

Founded by former Google Brain researcher Jeremy Nixon, Infinity.inc develops the universal software execution layer for AI hardware. By replacing manual software engineering with autonomous AI research agents, Infinity enables enterprises and chipmakers to deploy high-throughput AI inference on any silicon.