Benchmark Almanac

GPT-Image-2

GPT-Image-2 ranks first on 4 of the 14 benchmarks with five or more models scored.

14
Benchmarks with a score
4
Ranked first, out of five or more models
9
In the top three
OpenAI

GPT-Image-2

Closed weights

No Epoch Capabilities Index score yet
  • RISEBench++ 47.8% #3 /21
  • RISEBench 49.4% #2 /13
  • PaintBench 16.3% #1 /7

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Vision, audio and robotics

BenchmarkScoreRankSource
DesignArena Logo 1,365 #1 of 44 Official leaderboard
DesignArena Graphic Design 1,435 #2 of 43 Official leaderboard
UltraText Bench 99.3 #1 of 16 Official leaderboard
LMArena Text-to-Image Arena 1,382 #5 of 78 Official leaderboard
Qwen-Image-Bench 64.7 #1 of 12 Official leaderboard
LMArena Image Edit Arena 1,461 #5 of 58 Official leaderboard
DesignArena Image 1,344 #5 of 46 Official leaderboard
RISEBench++ 47.8% #3 of 21 Official leaderboard
PaintBench 16.3% #1 of 7 Official leaderboard
RISEBench 49.4% #2 of 13 Official leaderboard
DesignArena Image Editing 1,315 #5 of 21 Official leaderboard
VBVR-Bench 0.51 #2 of 8 Official leaderboard
VBVR-Pro 0.51 #2 of 8 Official leaderboard
StudyArena Image 1,216 #6 of 23 Official leaderboard