Horangi 4 - KoBALT-700 (Semantics): leaderboard
Metric: Accuracy (%). Source: horangi.ai. 105 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) (High) | 91 |
| 2 | Claude Fable 5 (High) | 87 |
| 3 | Gemini 3 Pro (Preview) (High) | 85 |
| 4 | GPT-5.6 Sol (Max) | 84 |
| 5 | GPT-5.6 Terra (Max) | 84 |
| 6 | GPT-5.5 (xHigh) | 83 |
| 7 | Gemini 3.5 Flash | 82.83 |
| 8 | GLM-5 | 82 |
| 9 | Qwen 3.7 Max | 81.25 |
| 10 | Grok 4.5 | 80.61 |
| 11 | GPT-5 | 80 |
| 12 | GPT-5.1 | 80 |
| 13 | Gemini 3 Flash (Preview) | 80 |
| 14 | GPT-5.6 Luna (Max) | 80 |
| 15 | GPT-5.2 (xHigh) | 80 |
Interactive version: theaggregate.ai/benchmark?slug=horangi-4-kobalt-700-semantics · How It Works · Data refreshed daily, snapshot 2026-09-19.