EvaOmni - Complex Reasoning - Answer Accuracy: leaderboard
EvaOmni (2026):1027questions from392videos,1FPS and480px long side with audio; shared prompt. Open non-MiMo models use recommended card configurations on8H20GPUs; MiMo and closed models use official APIs. Coverage/bottleneck grounding score correct answers against annotated audio and visual intervals; expert reference uses replay and is retained separately. Score (%).
Metric: Score (%). Source: arxiv.org. Saturation forecast: Around January 2027. 24 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.7 Flash | 79.1 | #22 |
| 2 | Gemini 3.5 Flash | 77.8 | #61 |
| 3 | Gemini 3.1 Pro (Preview) | 76.5 | #60 |
| 4 | Gemini 3.8 Flash | 74.5 | #24 |
| 5 | Gemini 3.6 Flash | 73.2 | #51 |
| 6 | Gemini 2.5 Pro | 72.5 | #154 |
| 7 | Gemini 4 Argon | 72.5 | #8 |
| 8 | Seed 2.0 Lite | 62.7 | #146 |
| 9 | Qwen3.8 Omni Flash | 56.2 | #33 |
| 10 | MiMo-V2.6-Pro | 49.7 | #38 |
| 11 | MiMo-V2.6-Flash | 45.8 | #73 |
| 12 | MiMo-V2.5 | 45.1 | #197 |
| 13 | Gemini 3.5 Flash Lite | 42.5 | #159 |
| 14 | Qwen 3.5 Omni Plus | 42.5 | #98 |
| 15 | Gemini 2.5 Flash | 41.8 | #246 |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=evaomni-complex-reasoning-answer-accuracy · How It Works · Data refreshed daily, snapshot 2026-10-22.