SciExam-ENSO - Composite: leaderboard
Metric: Composite score (0–1), final submitted model, six-hour budget. Source: github.com. Saturation forecast: Around January 2028. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5.1 | 0.52 |
| 2 | Kimi K3 | 0.48 |
| 3 | GPT-5.6 Sol | 0.44 |
| 4 | GPT-6 Astra | 0.44 |
| 5 | Claude Opus 5 | 0.44 |
| 6 | GPT-5.5 | 0.4 |
| 7 | Gemini 3.8 Flash | 0.37 |
| 8 | DeepSeek V4.1 Flash | 0.35 |
| 9 | DeepSeek V4 Pro | 0.34 |
| 10 | MiniMax-M3 | 0.22 |
| 11 | GLM-5.2 | 0.19 |
| 12 | Qwen 3 Max | 0.05 |
Interactive version: theaggregate.ai/benchmark?slug=sciexam-enso-composite · How It Works · Data refreshed daily, snapshot 2026-10-09.