LongShOTBench - Visual: leaderboard
LongShOTBench (arXiv:2512.16978) Visual modality slice: weighted criterion-level rubric score (0-100%) on visually grounded questions over long videos; higher is better.
Metric: Weighted Rubric Score (%). Source: longshot.cvmbzuai.com. Saturation forecast: Around 2029. 102 models tracked. Updated 21 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Qwen 3 Omni 30B A3B (Thinking) | 62.53 | ||
| 2 | Gemini 3.1 Pro (Preview) | 53.2 | #99 | |
| 3 | Voxtral-Mini-3B-2507 | 48.79 | #3682 | |
| 4 | Voxtral-Small-24B-2507 | 47.23 | #1246 | |
| 5 | Qwen 3 VL 235B A22B (Thinking) | 46.19 | #459 | |
| 6 | Qwen 3 VL 32B (Thinking) | 44.86 | #573 | |
| 7 | Vero-Qwen3I-8B | 44.22 | ||
| 8 | Kimi K2.6 | 43.14 | #192 | |
| 9 | Qwen 3 VL 30B A3B (Thinking) | 41.77 | #719 | |
| 10 | Qwen 3 VL 235B A22B Instruct | 40.52 | #532 | |
| 11 | Gemma 3 27B (IT) | 40.29 | #1324 | |
| 12 | Nemotron 3 Nano Omni 30B A3B | Reasoning | 40 | #1412 |
| 13 | Qwen 3 VL 8B (Thinking) | 39.89 | #910 | |
| 14 | Intern-S1-mini | 39.72 | #1178 | |
| 15 | Qwen 3 VL 32B Instruct | 38.47 | #606 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=longshotbench-visual · How It Works · Data refreshed daily, snapshot 2026-11-21.