ParseBench - Semantic Formatting: leaderboard
Metric: Semantic Formatting Score. Source: parsebench.ai. Saturation forecast: Around February 2028. 78 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GLM-5.3 Flash | 79.49 |
| 2 | Claude Opus 5.5 | 77.04 |
| 3 | Claude Fable 5.1 | 76.52 |
| 4 | Gemini 3.8 Flash (High) | 76.36 |
| 5 | DeepSeek V4.1 Flash (High) | 73.48 |
| 6 | Claude Fable 5 | 72.62 |
| 7 | Claude Opus 4.8 | 71.38 |
| 8 | Claude Opus 4.7 | 69.42 |
| 9 | Gemma 4 31B (IT) | 69.3 |
| 10 | Gemini 3 Flash (High) | 68.31 |
| 11 | Gemini 3.5 Flash (Minimal) | 68.09 |
| 12 | GPT-6 Sol (Medium) | 67.88 |
| 13 | GPT-6 Sol (Non-reasoning) | 67.32 |
| 14 | Gemini 3.8 Flash (Low) | 66.99 |
| 15 | Gemma 4 26B A4B (IT) | 65.1 |
Interactive version: theaggregate.ai/benchmark?slug=parsebench-semantic-formatting · How It Works · Data refreshed daily, snapshot 2026-09-24.