MUN-lang (Zero-Shot): leaderboard
Metric: Win rate (%; share of MUN-lang items where a GPT-4o judge, AlpacaEval-style, prefers the model's explanation (explain how an ordinary-looking image leads to an unusual outcome) over the human-written, GPT-4o-refined reference; zero-shot). Source: arxiv.org. Saturation forecast: Estimated already saturated. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 VL 7B Instruct | 42.2 |
| 2 | Phi-4 Multimodal Instruct | 35.7 |
| 3 | Qwen 2 VL 7B Instruct | 34.9 |
| 4 | Gemma 3 4B (IT) | 25.7 |
Interactive version: theaggregate.ai/benchmark?slug=mun-lang-zero-shot · How It Works · Data refreshed daily, snapshot 2026-09-26.