Mk1.5 real-time multimodal reasoning model ships with 2-5x lower latency than Mk1

AkshatS07 · x · 2026-09-25

Mk1.5 is a multimodal reasoning model built for real-time, physical-world use cases requiring fast decisions across multimodal inputs. The team claims accuracy at or above Mk1 while cutting latency 2-5x, unlocking new real-time applications.

Related event: Mk1.5 Real-Time Multimodal Reasoning Model Launches with 2-5x Lower Latency(4 posts)→

Original post →

More from Models

Models channel →