TECCI: leaderboard

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

#ModelScore
1Nano Banana Pro22.3
2Grok Imagine Pro20
3Nano Banana 2 (Gemini 3.1 Flash Image Preview)19.7
4Seedream 5.0 Lite16.7
5GPT Image 1.55.7

Interactive version: theaggregate.ai/benchmark?slug=tecci · How It Works · Data refreshed daily, snapshot 2026-09-26.