BOTTLED - ESCI: leaderboard
Metric: Macro F1 (0-1). Source: arxiv.org. Saturation forecast: Around June 2028. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5 | 0.5 |
| 2 | GLM-5.3 | 0.5 |
| 3 | GLM-5.3 Flash | 0.47 |
| 4 | Qwen 3.8 Max | 0.44 |
| 5 | Qwen 3.8 Flash | 0.42 |
| 6 | Gemini 3.1 Pro (Preview) | 0.38 |
| 7 | Claude Sonnet 5 | 0.35 |
| 8 | GPT-5.6 Terra | 0.27 |
| 9 | Gemini 3.1 Flash Lite | 0.25 |
| 10 | GPT-5.6 Sol | 0.2 |
Interactive version: theaggregate.ai/benchmark?slug=bottled-esci · How It Works · Data refreshed daily, snapshot 2026-10-09.