GB/T-Bench - Direct Review DMTR@9: leaderboard
Metric: Documents with at least 9 correct diagnoses (%). Source: arxiv.org. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 39.34 |
| 2 | GPT-5.6 Sol | 36.96 |
| 3 | Claude Fable 5 | 32.61 |
| 4 | GPT-5.4 | 25 |
| 5 | Qwen 3.7 Plus | 18.85 |
| 6 | Gemini 3.5 Flash | 17.21 |
| 7 | Claude Opus 4.8 | 14.34 |
| 8 | DeepSeek V4 Pro | 12.3 |
| 9 | GPT-5.3 | 8.81 |
| 10 | Qwen 3 235B A22B 2507 Instruct | 8.2 |
| 11 | DeepSeek V4 Flash | 7.38 |
| 12 | GPT-5.4 Mini | 3.89 |
| 13 | MiniMax-M2.7 | 2.87 |
| 14 | Qwen 2.5 VL 72B Instruct | 2.46 |
Interactive version: theaggregate.ai/benchmark?slug=gb-t-bench-direct-review-dmtr-9 · How It Works · Data refreshed daily, snapshot 2026-09-19.