Chinese Classical Bench - Fill-In Exact: leaderboard
Metric: Score (%). Source: huggingface.co. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 (Thinking) | 88 |
| 2 | Claude Opus 4.7 | 86 |
| 3 | Claude Sonnet 4.6 | 70 |
| 4 | MiniMax-M2.1 | 63 |
| 5 | MiniMax-M2.5 | 59 |
| 6 | DeepSeek V3.2 | 55 |
| 7 | Qwen3 Coder Next | 52 |
| 8 | GLM-5 | 45 |
| 9 | Qwen 3.5 35B A3B | 38 |
| 10 | Claude Haiku 4.5 (20251001) | 35 |
Interactive version: theaggregate.ai/benchmark?slug=chinese-classical-bench-fill-in-exact · How It Works · Data refreshed daily, snapshot 2026-09-05.