Claim: Kimi-K3 is 'chronically undertrained,' casting doubt on Moonshot's training budget

scaling01 · x · 2026-09-17

AI commentator scaling01 asserts that Moonshot AI's Kimi-K3 is "chronically undertrained," suggesting the flagship model falls short of its architectural potential due to insufficient training compute. The claim comes without supporting benchmarks or data, so it should be treated as an unverified take until further evals or official details emerge.

Original post →

More from Models

Models channel →