
TitanML is an optimization and compression platform that allows customers to achieve best-in-class model throughput, latency, and accuracy.

TitanML is an optimization and compression platform that allows customers to achieve best-in-class model throughput, latency, and accuracy.
Headquarters: London
Product: Self-hosted inference platform (Doubleword/Titan Takeoff) for LLMs and VLMs
Employees: 15
Total funding: USD 14,740,000
Rebrand: Now operating as Doubleword (formerly TitanML)
Model serving, inference optimization, compression, and private/self-hosted deployment for large language and vision models.
Enterprise AI / ML infrastructure
USD 2,800,000
USD 12,000,000
“K5 Global; Clément Delangue”