METR Q4 2026 TH2 - Dominion-like Game: leaderboard

METR's Q4 2026 capability update (20 November 2026): mean score from 0 to 1 on one task of TH2, METR's new long-task suite. The agent plays a deckbuilding card game modelled on Dominion. Higher is better.

Metric: Mean Score (0-1). Source: metr.org. Saturation forecast: Around February 2027. 15 models tracked. Updated 21 Nov 2026.

Top models

#ModelScoreOverall rank
1GPT-6 Astra1#10
2Claude Opus 5.50.9326#9
3GPT-5.50.8883#54
4Claude Fable 50.8182#33
5Claude Opus 50.7307#26
6Claude Fable 5.10.6242#20
7Claude Opus 4.80.2837#65
8GPT-5.40.2217#135
9Claude Opus 4.70.2133#85
10Claude Opus 4.60.1771#113
11Claude Opus 4.50.1195#160
12GPT-50.0803#178
13GPT-4o0#751
14GPT-5.6 Sol0#37
15O10#403

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=metr-q4-2026-th2-dominion-like-game · How It Works · Data refreshed daily, snapshot 2026-11-21.