Reasoning on HLE Verified (Gold)
60Accuracy (HLE Verified Gold)Metacognitive Harness
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Metacognitive HarnessBackbone=Sonnet-4.62026.05 | 60 | 12 | |
| GPT-5.2Reasoning Setting=High, Leaderboard Rank=Top-12026.05 | 52.5 | — | |
| Opus-4.6Leaderboard Rank=Top-22026.05 | 50.2 | — | |
| Gemini-3-ProLeaderboard Rank=Top-32026.05 | 48.9 | — | |
| Sonnet-4.6Evaluation Mode=Single-pass (Pass@1)2026.05 | 48 | — |