Kimi K3.1 Accused of Benchmark Contamination on TerminalBench

Users including TilmanRa and @JSchaeff3r allege Kimi K3.1's TerminalBench lead stems from knowledge leakage of hidden test sets, noting suspiciously strong performance on the hardest tasks and recall of undisclosed test rules.

2026-11-03 ~ 2026-11-03 · 2 related posts

• TerminalBench 4.0 questioned: Kimi K3.1's top score may leak hidden tests, per new critique — teortaxesTex · 2026-11-03
• Kimi K3.1 accused of recalling hidden Terminal-Bench test rules it was never given — maksym_andr · 2026-11-03