Horangi 4 - GLP - Mathematical Reasoning: leaderboard
Metric: Score (%). Source: horangi.ai. 105 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5 (High) | 98.5 |
| 2 | Grok 4.20 | 98 |
| 3 | GPT-5.4 Nano (xHigh) | 98 |
| 4 | GPT-5.4 (xHigh) | 97.5 |
| 5 | Claude Opus 4.8 (High) | 97.5 |
| 6 | Grok 4.5 | 97.47 |
| 7 | GPT-5.2 (xHigh) | 97 |
| 8 | Qwen 3.7 Max | 96.83 |
| 9 | GPT-5.5 (xHigh) | 96.5 |
| 10 | Gemini 3 Flash (Preview) | 96.33 |
| 11 | GPT-5.4 Mini (xHigh) | 96.33 |
| 12 | Gemini 3 Pro (Preview) (High) | 96.33 |
| 13 | A.X-K1 (Thinking) | 96 |
| 14 | DeepSeek V3.2 (High) | 95.89 |
| 15 | Gemini 3.1 Pro (Preview) (High) | 95.83 |
Interactive version: theaggregate.ai/benchmark?slug=horangi-4-glp-mathematical-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-19.