Equational Theories - Normal (Default reasoning): leaderboard
Metric: Strict F1 (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 25 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 | 97.98 |
| 2 | Gemini 3.1 Pro (Preview) | 97.81 |
| 3 | Qwen 3.5 397B A17B | 97.37 |
| 4 | Grok 4.1 Fast | 96.87 |
| 5 | Qwen 3.5 122B A10B | 96.79 |
| 6 | Qwen 3.5 27B | 94.97 |
| 7 | Gemini 3 Flash (Preview) | 94.1 |
| 8 | Kimi K2.5 | 93.59 |
| 9 | GPT-5 Mini | 93.44 |
| 10 | GPT-OSS-120B | 92.59 |
| 11 | GLM-5 | 91.73 |
| 12 | Seed 2.0 Lite | 90.52 |
| 13 | Step 3.5 Flash | 83.54 |
| 14 | Claude Sonnet 4.6 | 83 |
| 15 | MiniMax-M2.5 | 79.86 |
Interactive version: theaggregate.ai/benchmark?slug=equational-theories-normal-default-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-09.