SPD-Faith Bench - Difference Quantity Recall: leaderboard
Metric: Share (%) of pairs for which the model reports exactly the true number of differences, on the 1,000 multi-difference image pairs of SPD-Faith Bench (COCO 2017 images with two to five controlled edits each: colour change, object removal or position change, planned with an LLM, realized with LaMa inpainting and verified by human annotators); zero-shot chain-of-thought difference reasoning; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GLM-4.5V | 67 | #339 |
| 2 | Qwen 2.5 VL 7B Instruct | 65 | #643 |
| 3 | GPT-4o | 62.2 | #333 |
| 4 | Qwen 2.5 VL 72B Instruct | 61.6 | #364 |
| 5 | Gemini 2.5 Pro | 57.3 | #145 |
| 6 | Claude Haiku 4.5 | 54.2 | #271 |
| 7 | MiniCPM-V-2.6 | 49.6 | #825 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=spd-faith-bench-difference-quantity-recall · How It Works · Data refreshed daily, snapshot 2026-10-11.