Benchmark
Arena ELO Image Edit - Multi Image Edit Agentic
Agentic benchmarks measure how well models carry out tasks that take several steps. Higher scores are better.
All agentic benchmarksScores from LMArena data.
- Models measured
- 19
- Scores recorded
- 19
- Sources
- 1
- IQ weight
- 1.09
Score distribution
- 1,357.2
- Mean
- 1,367.4
- Median
- 103.9
- Std dev
- 1,221.5
- 10th percentile
- 1,482.7
- 90th percentile
- 1,170.3
- Min
This benchmark has a weight of 1.09 for image-generation models, measured over 17 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | GPT-Image-2.5 Sunburst· LMArena data | 1,549.9 |
| 2 | GPT Image 2Medium· LMArena data | 1,491.2 |
| 3 | GPT-Image-2.5 Flare· LMArena data | 1,482.7 |
| 4 | GPT-Image-2.6 FlareMedium· LMArena data | 1,470.6 |
| 5 | Seedream 5.0 Pro· LMArena data | 1,424.5 |
| 6 | Muse Image· LMArena data | 1,408.7 |
| 7 | Krea 3 Ultra· LMArena data | 1,402.7 |
| 8 | Gemini 3.1 Flash Imagenano-banana-2, Web Search· LMArena data | 1,391 |
| 9 | Gemini 3 Pro Imagenano-banana-pro, 2K· LMArena data | 1,389.7 |
| 10 | ChatGPT Image LatestHigh Fidelity, 20251216· LMArena data | 1,367.4 |
| 11 | GPT Image 1.5High Fidelity· LMArena data | 1,358.5 |
| 12 | MiniMax H3-Image· LMArena data | 1,351.8 |
| 13 | Ideogram V4.5· LMArena data | 1,319.7 |
| 14 | Doubao Seedream 4.5· LMArena data | 1,280.6 |
| 15 | UNI-1.1· LMArena data | 1,276.4 |
| 16 | Seedream 5.0 Lite· LMArena data | 1,248.5 |
| 17 | Gemini 2.5 Flash Imagenano-banana· LMArena data | 1,221.5 |
| 18 | FLUX.2 [pro]· LMArena data | 1,181.5 |
| 19 | Wan 2.6 Image· LMArena data | 1,170.3 |
Data credits: LMArena, CC BY 4.0 (licence)