version history

Benchmark history

Every benchmark run that was archived, newest first. The run marked live is the one the rest of this site is built from right now. Open any version to see its leaderboard as it stood, frozen.

merge: vertex/gemini-3.7-flash livebench v1.19.0rubric v1.5.0betajudged

1.19.0__20260823-201358

45 models · 10 seeds · 100 HP start · run Aug 23, 2026

view snapshot
merge: anthropic/claude-opus-5bench v1.19.0rubric v1.5.0betajudged

1.19.0__20260724-175531

44 models · 10 seeds · 100 HP start · run Jul 24, 2026

view snapshot
merge: xai/grok-4.5bench v1.19.0rubric v1.5.0betajudged

1.19.0__20260721-210318

43 models · 10 seeds · 100 HP start · run Jul 21, 2026

view snapshot

1.19.0__20260721-203000

42 models · 10 seeds · 100 HP start · run Jul 21, 2026

view snapshot
merge: moonshot/kimi-k3bench v1.19.0rubric v1.5.0betajudged

1.19.0__20260717-172246

40 models · 10 seeds · 100 HP start · run Jul 17, 2026

view snapshot

1.19.0__20260715-050626

39 models · 10 seeds · 100 HP start · run Jul 15, 2026

view snapshot

1.19.0__20260715-001313

40 models · 1 seed · 100 HP start · run Jul 15, 2026

view snapshot

1.18.0__20260713-235438

40 models · 1 seed · 100 HP start · run Jul 13, 2026

view snapshot

1.17.0__20260711-193641

35 models · 5 seeds · 100 HP start · run Jul 11, 2026

view snapshot

1.14.0__20260711-153309

35 models · 5 seeds · 100 HP start · run Jul 11, 2026

view snapshot

1.12.0__20260711-104416

36 models · 5 seeds · 100 HP start · run Jul 11, 2026

view snapshot

1.9.0__20260711-002344

36 models · 5 seeds · 100 HP start · run Jul 11, 2026

view snapshot

1.9.0__20260710-224030

36 models · 2 seeds · 100 HP start · run Jul 10, 2026

view snapshot

1.8.0__20260710-202759

36 models · 3 seeds · 100 HP start · run Jul 10, 2026

view snapshot