METR Q4 2026 TH2 - Judge Run (Older Models' Tasks): 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 grades another agent's attempt at building an evaluation task, using a rubric, and is scored on agreement with METR's human grader. This half uses tasks built by models released before 9 June 2026. 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
1Claude Opus 5.50.8236#9
2Claude Fable 5.10.7327#20
3GPT-6 Astra0.6694#10
4Claude Opus 50.6283#26
5Claude Opus 4.70.6186#85
6GPT-5.6 Sol0.5953#37
7Claude Opus 4.80.5679#65
8GPT-5.40.5068#135
9Claude Fable 50.5031#33
10Claude Opus 4.60.4997#113
11GPT-5.50.4549#54
12Claude Opus 4.50.4162#160
13GPT-50.2544#178
14O10.1716#403
15GPT-4o0.0308#751

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-judge-run-older-models-tasks · How It Works · Data refreshed daily, snapshot 2026-11-21.