Theo mocks claims that open-weight models match Opus 5.5 after Artificial Analysis’s reranking; replies question the ranking itself. You came for a benchmark and got jury selection.

Frame from Beamsters’ publicly posted animation comparing Artificial Analysis index v4.3.2 with v5, quoted by Theo; third-party visualization, not our independent model test.

Published 28 Oct 2026 · 22:39 UTC

Theo Browne and replies; quoted Beamsters visualization ↗ · Source dated 28 Oct 2026

Theo mocks claims that open-weight models match Opus 5.5 after Artificial Analysis’s reranking; replies question the ranking itself. You came for a benchmark and got jury selection. · ournews.ai