TECCI - Instruction Following: leaderboard

Metric: Instruction-following human success rate (%; five trained raters score each edit 1-5 and a criterion succeeds when the mean is at least 4.5; 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
1Grok Imagine Pro46.9
2Nano Banana Pro46
3Nano Banana 2 (Gemini 3.1 Flash Image Preview)45.6
4GPT Image 1.542
5Seedream 5.0 Lite41.4

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