Vector Eval - MMMU-MC: leaderboard
Vector Institute Inspect Evals run of MMMU multiple-choice tasks, evaluating multimodal college-level subject reasoning with image and text inputs.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O1 | 80.64 | #183 |
| 2 | Claude 3.5 Sonnet | 64.82 | #350 |
| 3 | Gemini 1.5 Pro | 63.05 | #445 |
| 4 | GPT-4o | 59.03 | #343 |
| 5 | Gemini 1.5 Flash | 57.02 | #618 |
| 6 | GPT-4o Mini | 53.96 | #599 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=vector-eval-mmmu-mc · How It Works · Data refreshed daily, snapshot 2026-10-20.