GB/T-Bench - Diagnosis Recall - Document Structure: leaderboard
GB/T-Bench (arXiv 2608.06312v1): review of 488 Chinese national standard documents seeded with 7,306 errors across 25 error types, by direct LLM review without the authors' multi-agent reviewer. Exact diagnosis recall for errors in the document-structure review dimension, such as section hierarchy and headings, percent (Table 3). Higher is better.
Metric: Diagnosis recall (%; exact section, dimension and error-type match). Source: arxiv.org. Saturation forecast: Around February 2027. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol | 75.81 | #22 |
| 2 | GPT-5.5 | 71.94 | #33 |
| 3 | Claude Fable 5 | 65.08 | #20 |
| 4 | GPT-5.4 | 55.87 | #83 |
| 5 | Qwen 3.7 Plus | 54.41 | #98 |
| 6 | DeepSeek V4 Pro | 50.33 | #112 |
| 7 | Gemini 3.5 Flash | 49.87 | #66 |
| 8 | GPT-5.3 | 48.26 | #104 |
| 9 | DeepSeek V4 Flash | 41.81 | #155 |
| 10 | Claude Opus 4.8 | 35.57 | #39 |
| 11 | GPT-5.4 Mini | 31.99 | #228 |
| 12 | MiniMax-M2.7 | 29.22 | #270 |
| 13 | Qwen 2.5 VL 72B Instruct | 25.34 | #383 |
| 14 | Qwen 3 235B A22B 2507 Instruct | 24.74 | #315 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=gb-t-bench-diagnosis-recall-document-structure · How It Works · Data refreshed daily, snapshot 2026-11-05.