HoloCount - Material Attribution: leaderboard

Metric: Accuracy (%; exact-match accuracy of the single integer count a multimodal model returns zero-shot for an image and a counting question, with a strict integer-only system prompt; counting objects of a stated material (101 questions); higher is better). Source: arxiv.org. Saturation forecast: Around December 2026. 30 models tracked.

Top models

#ModelScore
1Qwen 3.5 27B85.1
2Gemini 3.1 Pro (Preview)84.2
3Kimi K2.683.2
4Qwen 3.5 397B A17B83.2
5Qwen 3.5 35B A3B83.2
6Qwen 3.5 122B A10B82.2
7Qwen 3.5 9B80.2
8Gemini 3 Flash (Preview)77.2
9Qwen 3.5 4B77.2
10Gemini 2.5 Pro76.2
11GPT-5.576.2
12Kimi K2.575.2
13Claude Opus 4.875.2
14Qwen 2.5 VL 72B Instruct74.3
15Claude Opus 4.772.3

Interactive version: theaggregate.ai/benchmark?slug=holocount-material-attribution · How It Works · Data refreshed daily, snapshot 2026-09-29.