GB/T-Bench - Diagnosis Recall - Scope Consistency: leaderboard
Metric: Diagnosis recall (%; exact section, dimension and error-type match). Source: arxiv.org. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 81.81 |
| 2 | GPT-5.6 Sol | 75.96 |
| 3 | Claude Fable 5 | 70.87 |
| 4 | GPT-5.4 | 69.65 |
| 5 | Qwen 3.7 Plus | 67 |
| 6 | Gemini 3.5 Flash | 60.88 |
| 7 | DeepSeek V4 Pro | 60.33 |
| 8 | GPT-5.3 | 54.3 |
| 9 | DeepSeek V4 Flash | 53.38 |
| 10 | GPT-5.4 Mini | 43.33 |
| 11 | Claude Opus 4.8 | 41.96 |
| 12 | Qwen 2.5 VL 72B Instruct | 38.39 |
| 13 | Qwen 3 235B A22B 2507 Instruct | 36.11 |
| 14 | MiniMax-M2.7 | 33.46 |
Interactive version: theaggregate.ai/benchmark?slug=gb-t-bench-diagnosis-recall-scope-consistency · How It Works · Data refreshed daily, snapshot 2026-09-19.