SFI-Bench - Operation Planning: leaderboard
Operation planning questions, on how to operate appliances and 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 operation planning 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 | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | LLaVA-OneVision-72B | 58.4 | #1771 | |
| 2 | LLaVA-Video-72B | 58.4 | #1193 | |
| 3 | Qwen 3 VL 8B Instruct | 54.9 | #1029 | |
| 4 | Qwen 3 VL 32B Instruct | 53.1 | #606 | |
| 5 | InternVL3.5-8B | 53.1 | #1604 | |
| 6 | Qwen 3 VL 235B A22B Instruct | 53 | #532 | |
| 7 | InternVL3.5-14B | 52.2 | #1504 | |
| 8 | Qwen 3 VL 235B A22B (Thinking) | 51.3 | #459 | |
| 9 | InternVL3.5-30B-A3B | 51.3 | #1267 | |
| 10 | Qwen 3 VL Plus | 50.4 | #512 | |
| 11 | LLaVA-OneVision-7B | 49.9 | #2981 | |
| 12 | Qwen 3 VL 30B A3B Instruct | 49.6 | #895 | |
| 13 | Qwen 3 VL 32B (Thinking) | 42.5 | #573 | |
| 14 | GLM-4.5V | Thinking | 42.5 | #754 |
| 15 | Qwen 3 VL 8B (Thinking) | 40.7 | #910 |
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-operation-planning · How It Works · Data refreshed daily, snapshot 2026-11-21.