Old Zhang's AI Learning

Oct 9, 2026 · Artificial Intelligence

Liquid AI's d1-3B Decides in 8ms: Open-Weight 3B Decision Model Released

Liquid AI releases two open-weight decision models, d1-3B and d1-omni-600M, which use single-forward-pass inference to achieve 8ms latency on RTX 4090, eliminating token generation for classification, routing, and scoring tasks with zero output tokens.