Liquid AI is an MIT spinout developing foundation models based on liquid neural networks — a continuous-time architecture that originated in Ramin Hasani's MIT CSAIL research on liquid time-constant networks and Mathias Lechner's work on continuous-time neural nets. The architecture targets dramatically smaller models with competitive quality and far lower inference compute than transformer baselines, making it attractive for on-device and edge deployments. Liquid AI raised a $250M Series A in December 2024 at a $2.35B post-money valuation, led by AMD Ventures with participation from existing investors. Total raised: ~$297M across Seed and Series A.

Company Profile

Founders & Key People

Investors

Products

• Liquid Foundation Models (LFMs)
• Liquid Edge AI
• Liquid Vision

Competitors

AIFoundation ModelsMIT spinoutLiquid Neural Networks

Private-company numbers are not real-time. Reflects publicly disclosed valuations from press releases, news reports, and tender offers as of 2024-Q4. Refreshed quarterly.