Chronicles-OCR - Archaic Character Recognition: leaderboard
Metric: Exact-match accuracy (%) of the modern character given for an archaic symbol highlighted by a box on the image, on the archaic scripts (Oracle Bone, Bronze and Seal), the paper Average column over the per-script results; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 29 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K2.5 (Non-reasoning) | 27.1 |
| 2 | Seed 2.0 Pro (Non-reasoning) | 24.5 |
| 3 | Seed 2.0 Pro | 23.3 |
| 4 | Qwen 3.5 397B A17B (Non-reasoning) | 22.6 |
| 5 | Kimi K2.5 | 20.3 |
| 6 | Gemini 3.1 Pro (Preview) | 19.5 |
| 7 | Qwen 3 VL 235B A22B Instruct | 17.5 |
| 8 | Qwen 3 VL 8B Instruct | 17.3 |
| 9 | Seed 1.8 | 17.1 |
| 10 | Qwen 3.5 35B A3B (Non-reasoning) | 16.2 |
| 11 | Qwen 3 VL 235B A22B (Thinking) | 13.6 |
| 12 | Claude Opus 4.7 (Thinking) | 10 |
| 13 | Qwen 3 VL 8B (Thinking) | 9.1 |
| 14 | MiMo-V2-Omni | 8.6 |
| 15 | Gemini 2.5 Pro | 7.5 |
Interactive version: theaggregate.ai/benchmark?slug=chronicles-ocr-archaic-character-recognition · How It Works · Data refreshed daily, snapshot 2026-10-07.