PaleBlueDot AI said on 1 October it had completed a $200 million Series C led by ComputeCore at a $3.2 billion valuation, following a $150 million Series B announced in January.
Key points
- B Capital, an existing shareholder, joined ComputeCore and other investors in the Series C, PaleBlueDot AI said.
- The company said it had signed over $5 billion in customer contracts by the end of September 2026.
- Customers in the U.S. and Japan together accounted for more than half of monthly revenue, according to the company.
- PaleBlueDot AI said it would use the proceeds to expand compute capacity.
ComputeCore leads PaleBlueDot AI’s Series C
PaleBlueDot AI said B Capital, already a shareholder, participated alongside other global investors. The Palo Alto company said it was founded in 2024 and had announced its previous financing, a $150 million Series B, in January 2026.
The company said the new funding would pay for additional compute capacity. It said the expansion would give customers more choice over hardware and locations. That planned spending covers a business which PaleBlueDot AI describes as combining clusters it owns and operates with access to capacity supplied by other companies.
Over $5 billion in customer contracts
By the end of September 2026, PaleBlueDot AI had signed over $5 billion in customer contracts, according to the company. It also said customers from the U.S. and Japan together contributed more than half of its monthly revenue. The contract total covers agreements signed by that date; the geographic figure is the company’s account of where its monthly revenue came from.
PaleBlueDot AI said it diversified its data-centre capacity and customer base in 2026. Its clusters operate across multiple regions, the company said, while its customers span markets worldwide. The U.S. and Japan revenue figure gives a more specific account of where a majority of monthly income was generated.
Japan cluster and marketplace capacity
PaleBlueDot AI said it customises its dedicated GPU clusters for customers with large or specialised workloads. Through its marketplace, it offers on-demand compute from a global network of supply partners. The company also offers serverless inference to developers and enterprises. It calls the combination its Super Intelligence Infrastructure Platform.
In Japan, the company said its B300 cluster had earned Nvidia Exemplar Cloud status. PaleBlueDot AI said that designation assessed the cluster’s performance against Nvidia reference benchmarks. The Japan cluster sits within the part of its business that owns and operates capacity, alongside the marketplace through which customers can access partners’ supply.
Stephen Watts, PaleBlueDot AI’s CEO, said the company would broaden its customer base, focusing on frontier labs, Neolabs and enterprises in the U.S. He said it would also invest in its full-stack and go-to-market teams.
