Flapping Airplanes raises $1B Series A at reported $6B valuation

The round was announced nine months after Flapping Airplanes launched with a $180 million seed round from GV, Sequoia Capital and Index, TechCrunch reported. That seed valued the company at $1.5 billion, according to TechCrunch, so the new valuation would roughly quadruple it. SiliconANGLE, citing Bloomberg’s count, reported that investors have now put $1.18 billion into the company. Sequoia, GV and Kleiner Perkins returned for the Series A, along with new investors including Valor, Atreides, A*, Jump Trading, XTX Ventures and Nvidia, according to the company’s posts and the two reports.

Flapping Airplanes was founded by Aidan Smith and brothers Asher and Ben Spector. Smith, a Thiel Fellow, worked at Neuralink during college, according to TechCrunch. Asher Spector was a statistics PhD at Stanford, and Ben Spector is a Stanford computer science PhD student who founded the student-run startup accelerator Prod while at MIT, SiliconANGLE reported; it said Ben Spector is CEO and was lead author in 2024 on ThunderKittens, an open-source toolkit for writing fast GPU kernels. Index describes a team of a couple dozen people working in person in San Francisco. Mark Xu of Index is joining the board, SiliconANGLE reported.

The founders say their main technical result is what they call “the world’s most complete megakernel framework.” A kernel is a program that runs on a GPU; a megakernel bundles many tasks into one program. Flapping Airplanes says its framework can perform an entire training run, serve inference, play a video game, and do all three simultaneously in a single kernel, according to TechCrunch. The company also claims it can overlap memory-bound inference with compute-bound training, so a chip can keep working while one task waits for data and another performs math.

The aim is to lower the cost of training and running models. Asher Spector told TechCrunch shortly after the company launched that “a model that’s a million times more data efficient is probably a million times easier to put into the economy.” The company argues AI has barely scratched the surface of what GPUs can do and that intelligent systems can be built “without ingesting half the internet.” In its announcement on X, the company said its first year went into new GPU primitives that enable algorithms that could “dramatically reshape AI’s cost structure,” SiliconANGLE reported.

Index investors Mark Xu and Shardul Shah wrote that demand for AI will keep outrunning the supply of compute and data, and singled out robotics, trading and life sciences as markets that need more efficient AI, according to SiliconANGLE. Sequoia partner David Cahn, an investor, described the founders in a post earlier this year as building the “young person’s AGI lab,” TechCrunch reported. The new capital is intended to fund hiring and turn the lab’s data-efficiency goal into working systems, though no commercial product has been announced.