Nathan Lambert calls out Ethan Mollick for 'selectively ignoring' Kimi in model comparisons

natolambert · x · 2026-11-16

Nathan Lambert shared an Artificial Analysis Progress Index chart showing Kimi K3.1 Pro closing in on Anthropic's Fable, accusing Ethan Mollick of "selectively ignoring" Kimi in his comparisons. Mollick replied that Kimi is good but not Fable-level, describing it as brittle outside of coding tasks — a snapshot of the ongoing debate over whether frontier Chinese models get underweighted.

Original post →

More from Models

Models channel →