SFI-Bench - Layout Inference: leaderboard
Room layout inference questions from SFI-Bench (2026): four-option questions about egocentric videos of scanned indoor rooms, answered zero-shot from the video. Accuracy in percent; higher is better.
Metric: Accuracy (%) on the layout inference questions, four-option multiple choice on egocentric indoor videos (ARKitScenes and ScanNet++ scans), zero-shot with the same prompt templates, models at their default configurations; answered directly with no tools; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 24 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 86.8 | #99 |
| 2 | Gemini 2.5 Pro | 83.8 | #276 |
| 3 | GPT-5.4 | 83 | #135 |
| 4 | O4 Mini | 82.4 | #368 |
| 5 | GPT-5 | 81.5 | #178 |
| 6 | Gemini 3.1 Flash Lite | 81.3 | #371 |
| 7 | Qwen 3 VL 235B A22B Instruct | 78.8 | #532 |
| 8 | Qwen 3 VL 32B Instruct | 76.7 | #606 |
| 9 | Qwen 3 VL 32B (Thinking) | 75.7 | #573 |
| 10 | Qwen 3 VL 30B A3B Instruct | 75.5 | #895 |
| 11 | LLaVA-Video-72B | 75.2 | #1193 |
| 12 | Qwen 3 VL 30B A3B (Thinking) | 75 | #719 |
| 13 | Qwen 3 VL 235B A22B (Thinking) | 74 | #459 |
| 14 | InternVL3.5-30B-A3B | 74 | #1267 |
| 15 | Qwen 3 VL Plus | 73.6 | #512 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=sfi-bench-layout-inference · How It Works · Data refreshed daily, snapshot 2026-11-21.