SFI-Bench - Functional Association: leaderboard

Functional association questions, which link objects to what they are for, 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 functional association 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

#ModelSettingScoreOverall rank
1GPT-5.4High76.2#138
2GPT-575.3#185
3Gemini 3.1 Pro (Preview)73.2#102
4O4 Mini68.5#372
5Gemini 2.5 Pro65.5#280
6Qwen 3 VL Plus61.3#514
7Qwen 3 VL 235B A22B (Thinking)60.9#462
8LLaVA-OneVision-72B60.1#1782
9Qwen 3 VL 32B (Thinking)59.7#577
10Gemini 3.1 Flash Lite58.3#377
11LLaVA-Video-72B56.7#1211
12Qwen 3 VL 235B A22B Instruct55.5#541
13Qwen 3 VL 32B Instruct55.5#607
14Gemini 2.5 Flash52.1#466
15InternVL3.5-30B-A3B50#1269

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=sfi-bench-functional-association · How It Works · Data refreshed daily, snapshot 2026-11-22.