LongShOTBench - Core Perception: leaderboard
LongShOTBench (arXiv:2512.16978) Core Perception (CP) category breakdown: weighted criterion-level rubric score (0-100%) on fine-grained perceptual recognition across long omni-modal videos; higher is better.
Metric: Weighted Rubric Score (%). Source: longshot.cvmbzuai.com. Saturation forecast: Around 2030. 102 models tracked. Updated 21 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 Omni 30B A3B (Thinking) | 51.23 | |
| 2 | Gemini 3.1 Pro (Preview) | 44.67 | #99 |
| 3 | Voxtral-Mini-3B-2507 | 41.37 | #3682 |
| 4 | Vero-Qwen3I-8B | 41.08 | |
| 5 | Qwen 3 VL 235B A22B (Thinking) | 40.5 | #459 |
| 6 | Voxtral-Small-24B-2507 | 39.66 | #1246 |
| 7 | Qwen 3 VL 32B (Thinking) | 39.39 | #573 |
| 8 | Kimi K2.6 | 38.16 | #192 |
| 9 | Qwen 3 VL 235B A22B Instruct | 36.64 | #532 |
| 10 | Qwen 3 VL 30B A3B (Thinking) | 36.52 | #719 |
| 11 | Gemma 3 27B (IT) | 35.43 | #1324 |
| 12 | Qwen 3 VL 32B Instruct | 35.34 | #606 |
| 13 | Qwen 3 VL 8B (Thinking) | 35.11 | #910 |
| 14 | Intern-S1-mini | 34.47 | #1178 |
| 15 | Gemma 3 12B (IT) | 32.9 | #1887 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=longshotbench-core-perception · How It Works · Data refreshed daily, snapshot 2026-11-21.