LLM Stats (Vibe-Eval): leaderboard
269 challenging visual understanding prompts rated by human annotators, testing nuanced image comprehension beyond standard VQA.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around January 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 06-05) | 67.2 | #106 |
| 2 | Gemini 2.5 Pro | 65.6 | #152 |
| 3 | Gemini 2.5 Flash | 65.4 | #245 |
| 4 | Gemini 2.0 Flash | 56.3 | #342 |
| 5 | Gemini 1.5 Pro | 53.9 | #445 |
| 6 | Gemini 2.5 Flash Lite | 51.3 | #422 |
| 7 | Gemini 1.5 Flash | 48.9 | #618 |
| 8 | Gemini 1.5 Flash-8B | 40.9 | #819 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-vibe-eval · How It Works · Data refreshed daily, snapshot 2026-10-20.