METR Q4 2026 TH2 - Dominion-like Game (Budgeted): 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 Dominion-like deckbuilding game in the variant METR's data labels budgeted. 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-5.6 Sol1#37
2Claude Opus 51#26
3GPT-6 Astra1#10
4Claude Fable 5.11#20
5GPT-5.50.9481#54
6Claude Opus 4.80.8079#65
7Claude Opus 5.50.7262#9
8Claude Fable 50.4494#33
9Claude Opus 4.70.2781#85
10Claude Opus 4.60.1106#113
11Claude Opus 4.50.1065#160
12GPT-5.40.0908#135
13GPT-50.0533#178
14GPT-4o0#751
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-budgeted · How It Works · Data refreshed daily, snapshot 2026-11-21.