Visual Reasoning on R-Bench-V (Overall)
41.1AccuracyMetacognitive Harness
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Metacognitive HarnessBackbone=Sonnet-4.62026.05 | 41.1 | 5 | |
| Sonnet-4.6Evaluation Mode=Single-pass (Pass@1)2026.05 | 36.1 | — | |
| GPT-5-miniLeaderboard Rank=Top-12026.05 | 27.9 | — | |
| o3Leaderboard Rank=Top-22026.05 | 25.8 | — | |
| o4-miniLeaderboard Rank=Top-32026.05 | 20.9 | — |