NovGauge - Pairwise Novelty - Method Verified F1: leaderboard
Metric: Verified F1 (%). Source: arxiv.org. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 43.1 |
| 2 | GPT-5.4 Mini | 29.2 |
| 3 | Gemini 3.1 Pro (Preview) | 19.9 |
| 4 | Gemma 4 31B (IT) | 19.2 |
| 5 | Gemini 3.1 Flash Lite | 15.5 |
| 6 | Kimi K2.6 | 12.9 |
| 7 | GLM-5.1 | 9.4 |
| 8 | Claude Opus 4.7 | 9.1 |
| 9 | Qwen 3 32B | 8.9 |
| 10 | Claude Sonnet 4.6 | 8.6 |
| 11 | Qwen 3.6 Plus | 8.2 |
| 12 | Qwen 3.5 27B | 8.2 |
| 13 | DeepSeek V4 Pro | 6.7 |
| 14 | MiniMax-M2.7 | 5.5 |
| 15 | DeepSeek V4 Flash | 5.4 |
Interactive version: theaggregate.ai/benchmark?slug=novgauge-pairwise-novelty-method-verified-f1 · How It Works · Data refreshed daily, snapshot 2026-09-19.