BenchCAD 2.1 Core (OpenAI runs, Codex): leaderboard
Metric: Core score (0-100; BenchCAD's 0-1 score x 100) on the 100 held-out BenchCAD 2.1 Core cases, OpenAI's runs in Codex. Source: openai.com. Saturation forecast: Around March 2028. 4 models tracked. Updated 18 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-6.2 Astra | Max | 82.5 | #2 |
| 2 | GPT-6.2 Sol | Max | 73.1 | #10 |
| 3 | GPT-6 Astra | Max | 62.6 | #7 |
| 4 | GPT-6.1 Sol | Max | 52.6 | #12 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Carbon, #6 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=benchcad-2-1-core-openai-runs-codex · How It Works · Data refreshed daily, snapshot 2026-11-18.