DiffCap-Bench: leaderboard
Image Difference Captioning (IDC) generates natural language descriptions that precisely identify differences between two images, serving as a key benchmark for fine-grained change perception, cross-modal reasoning, and image editing data construction. However, existing benchmarks lack diversity and compositional complexity, and standard lexical-overlap metrics (e.g., BLEU, METEOR) fail to capture
Metric: F1* (self-reported). Source: benchmarklist.com. Saturation forecast: Around April 2027. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro | 83.1 | #80 |
| 2 | GPT-5.2 | 75.1 | #110 |
| 3 | Kimi K2.5 (Thinking) | 74.5 | #144 (Kimi K2.5) |
| 4 | Kimi K2.5 | 71.3 | #144 |
| 5 | Grok 4.20 | 69.6 | #175 |
| 6 | Qwen 3 VL 32B (Thinking) | 68.7 | #294 (Qwen 3 VL 32B) |
| 7 | Step3 VL 10B | 66.1 | #485 |
| 8 | Qwen 3 VL 8B (Thinking) | 65.5 | |
| 9 | Qwen 3 VL 32B Instruct | 63.8 | #283 |
| 10 | Qwen 3 VL 8B Instruct | 61.7 | #420 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=diffcap-bench · How It Works · Data refreshed daily, snapshot 2026-10-20.