METR Q4 2026 TH2 - AZsPCs Polygonal Areas (Human Percentile): leaderboard
METR's Q4 2026 capability update (20 November 2026): where a model's best 48-hour run on the past Al Zimmermann's Programming Contest "Polygonal Areas" ranks among that contest's 329 human entrants, who had about three months. 100 means beating every human entrant. One of the 15 contests in the TH2 suite.
Metric: Percentile Among Human Participants. Source: metr.org. Saturation forecast: Around February 2027. 10 models tracked. Updated 21 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 98.48 | #10 |
| 2 | Claude Opus 5.5 | 95.14 | #9 |
| 3 | Claude Fable 5.1 | 94.22 | #20 |
| 4 | GPT-5.6 Sol | 89.06 | #37 |
| 5 | Claude Haiku 5.5 | 87.23 | #75 |
| 6 | GPT-6 Luna | 86.32 | #105 |
| 7 | Claude Opus 4.7 | 76.29 | #85 |
| 8 | GPT-5.4 | 74.77 | #135 |
| 9 | Claude Opus 4.5 | 72.34 | #160 |
| 10 | GPT-5 | 60.79 | #178 |
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-azspcs-polygonal-areas-human-percentile · How It Works · Data refreshed daily, snapshot 2026-11-21.