# MambaAct: Efficient Mamba Action Experts for Flow-Matching Vision-Language-Action Models > Source: > Published: 2026-10-23 10:38:59+00:00 Vision-language-action (VLA) models have shown strong promise for generalist robotic manipulation, but their inference cost remains a major bottleneck for real-time control. This limitation is especially pronounced in flow-matching VLAs such as π_{0.5}, where the action expert is evaluated repeatedl