Flapping Airplanes raises $1B Series A at $6B valuation, nine months after seed round
The one-year-old startup’s $1B Series A values it at $6B, quadrupling its valuation from $1.5B nine months ago. Its megakernel framework aims to train large AI models with less data while keeping GPUs busier.
Flapping Airplanes, a one-year-old AI startup, announced on Wednesday, October 21, 2026, that it has raised a $1 billion Series A round co-led by Index and Lux at a reported $6 billion valuation — quadrupling its $1.5 billion valuation from just nine months earlier.
The round follows the company’s $180 million seed round from GV, Sequoia, and Index, which launched the startup earlier in 2026. Additional participants in the new Series A include returning backers Sequoia, GV, and Kleiner Perkins, alongside new investors Valor, Atreides, A*, and Nvidia.
Flapping Airplanes was founded by Aidan Smith and brothers Asher and Ben Spector. Smith is a former founder and Thiel fellow who worked at Neuralink during college. Asher Spector holds a statistics PhD from Stanford, and Ben Spector was a computer science PhD student at Stanford and founded Prod while at MIT. Prod is a five-year-old, student-run startup accelerator that Sequoia partner David Cahn — himself an investor in the company — described earlier this year as “one of the most powerful talent incubators in Silicon Valley.”
The company is focused on training large AI models using significantly less data than current methods require. Its core technology is what the founders describe as “the world’s most complete megakernel framework” — a system that bundles multiple GPU tasks into a single program, allowing training and inference workloads to run simultaneously so the chip spends less time idle. The founders argue that AI has “barely scratched the surface” of what GPUs can do.
“A model that’s a million times more data efficient is probably a million times easier to put into the economy,” Asher Spector told TechCrunch at the time of the company’s launch.
The fresh capital will be used to hire and advance the company’s goal of building AI systems that cost less to train and operate. If the approach proves out, it could suggest a path to capable AI models that do not depend on ingesting vast quantities of internet data.