SFI-Bench - Troubleshooting: leaderboard
Causal hypothesis and troubleshooting questions about devices seen in the video from SFI-Bench (2026): four-option questions about egocentric videos of scanned indoor rooms, answered without a web search tool. Accuracy in percent; higher is better.
Metric: Accuracy (%) on the causal hypothesis and troubleshooting 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; models without a search tool, answered offline; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 17 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | LLaVA-OneVision-72B | 61 | #1771 |
| 2 | Qwen 3 VL 235B A22B Instruct | 58.1 | #532 |
| 3 | InternVL3.5-14B | 54.7 | #1504 |
| 4 | Qwen 3 VL 32B Instruct | 54.6 | #606 |
| 5 | InternVL3.5-8B | 52.3 | #1604 |
| 6 | InternVL3.5-30B-A3B | 52.3 | #1267 |
| 7 | LLaVA-Video-72B | 50.9 | #1193 |
| 8 | LLaVA-OneVision-7B | 50.3 | #2981 |
| 9 | Qwen 3 VL 30B A3B (Thinking) | 50 | #719 |
| 10 | Qwen 3 VL 8B Instruct | 48.8 | #1029 |
| 11 | Qwen 3 VL 8B (Thinking) | 47.7 | #910 |
| 12 | Qwen 3 VL Plus | 47.7 | #512 |
| 13 | Qwen 3 VL 30B A3B Instruct | 45.3 | #895 |
| 14 | Qwen 3 VL 235B A22B (Thinking) | 45.3 | #459 |
| 15 | Qwen 3 VL 32B (Thinking) | 44.2 | #573 |
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-troubleshooting · How It Works · Data refreshed daily, snapshot 2026-11-21.