TECCI: leaderboard
Instruction-guided image editing on TECCI (Agrawal et al., 2026), newly collected images in seven categories picked for hard edits, judged by five trained raters on 1,315 sampled edits. Overall success rate in percent: the share of edits whose mean rating reaches 4.5 of 5 for instruction following, image consistency and visual quality at once; higher is better.
Metric: Overall human success rate (%; five trained raters score each edit 1-5 and a criterion succeeds when the mean is at least 4.5 on all three criteria; 1,315 sampled pairs: 265 of the Instruction Rich Challenge Set and 1,050 of the Gemini Generated Instruction Set). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Nano Banana Pro | 22.3 | |
| 2 | Grok Imagine Pro | 20 | |
| 3 | Nano Banana 2 (Gemini 3.1 Flash Image Preview) | 19.7 | |
| 4 | Seedream 5.0 Lite | 16.7 | |
| 5 | GPT Image 1.5 | 5.7 |
Interactive version: theaggregate.ai/benchmark?slug=tecci · How It Works · Data refreshed daily, snapshot 2026-11-12.