Skip to content

Benchmark

Arena ELO Text to Image - Cartoon

Higher scores are better.

Models measured
79
Scores recorded
86
Sources
1
IQ weight
1.40

Score distribution

1,151.1
Mean
1,142.3
Median
123.7
Std dev
1,006.5
10th percentile
1,329.8
90th percentile
883.2
Min
883.21,459.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.40 for image-generation models, measured over 68 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT-Image-2.5 Sunburst· LMArena data1,459.1
2GPT-Image-2.6 SunburstMedium· LMArena data1,447.1
3GPT-Image-2.5 Flare· LMArena data1,426
4MAI Image 2.7· LMArena data1,411.9
5GPT Image 2Medium· LMArena data1,404
6GPT-Image-2.6 FlareMedium· LMArena data1,401.8
7Grok Imagine Image 2.0· LMArena data1,352.9
8MAI-Image-2.6· LMArena data1,348.9
9Gemini Nano Banana 2.1· LMArena data1,329.8
10Reve 2.1· LMArena data1,312.1
11Grok Imagine Image 2.020260801· LMArena data1,300.2
12Qwen Image 3.0 Pro· LMArena data1,276
13Reve 2.0· LMArena data1,275.1
14Muse Image· LMArena data1,274.7
15Mai Image 2.5· LMArena data1,269.3
16Seedream 5.0 Pro· LMArena data1,261.7
17Gemini 3.1 Flash Imagenano-banana-2, Web Search· LMArena data1,260.8
18Gemini 3.1 Flash-Lite Imagenano-banana-2-lite· LMArena data1,256
19GPT Image 1.5High Fidelity· LMArena data1,243.7
20Gemini 3 Pro Imagenano-banana-pro, 2K· LMArena data1,243.2
21Gemini 3 Pro Imagenano-banana-pro· LMArena data1,236.4
22Qwen Image 2.1· LMArena data1,231.9
23Ideogram V4Quality· LMArena data1,193.7
24UNI-1.1 Max· LMArena data1,192.1
25MiniMax H3-Image· LMArena data1,192
26UNI-1.1· LMArena data1,190.1
27Qwen Image 2.0 Pro2026-06-22· LMArena data1,186.8
28Cosmos3-Super-Text2ImageAgentic· LMArena data1,185.1
29MAI-Image-2· LMArena data1,180.6
30Recraft V4.1 Utility Pro· LMArena data1,177.6
31Cosmos3-Super-Text2Image· LMArena data1,174.8
32Grok Imagine Image· LMArena data1,172.9
33FLUX.2 [max]· LMArena data1,167.5
34Hunyuan Image V3· LMArena data1,164.3
35FLUX.2 [flex]· LMArena data1,163.5
36Grok Imagine Image Pro· LMArena data1,161.6
37FLUX.2 [pro]· LMArena data1,159
38Doubao Seedream 4.5· LMArena data1,149.1
39Seedream 5.0 Lite· LMArena data1,147.6
40Doubao Seedream 4-02K· LMArena data1,147.5
41Wan 2.6· LMArena data1,147.1
42Gemini 2.5 Flash Imagenano-banana· LMArena data1,144.6
43FLUX.2 [dev]· LMArena data1,142.3
44Reve v1.5· LMArena data1,141.9
45Krea v2 Medium· LMArena data1,140.5
46Imagen 4 Ultragenerate-001· LMArena data1,139.5
47Krea 2 Turbo· LMArena data1,129.2
48Recraft V4.1 Pro· LMArena data1,125.9
49Qwen Image2512· LMArena data1,124.6
50Wan 2.5 Preview· LMArena data1,118.6

Showing 1–50 of 86 results.

Data credits: LMArena, CC BY 4.0 (licence)