Artificial Analysis Image Models (October 2026)

In October 2026, GPT-Image-2.5 Sunburst ranks first with 1,197, followed by GPT-Image-2.5 Flare (1,190) and GPT Image 2 (1,172). Top 20 of 137 candidates.

Data as of October 2026

#ModelProviderScore
1GPT-Image-2.5 SunburstOpenAI1,197
2GPT-Image-2.5 FlareOpenAI1,190
3GPT Image 2OpenAI1,172
4Grok Imagine Image 2.0xAI1,155
5MAI-Image-2.6Microsoft1,150
6Nano Banana 2 (Gemini 3.1 Flash Image Preview)Google1,125
7Muse ImageMeta1,114
8MAI-Image-2.6-FlashMicrosoft1,105
9MAI-Image-2.5Microsoft1,104
10Nano Banana Pro (Gemini 3 Pro Image)Google1,101
11MAI-Image-2.5-ProMicrosoft1,100
12Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)Google1,096
13Qwen-Image-3.0-ProAlibaba1,089
14Seedream 5.0 ProByteDance1,081
15Qwen-Image-3.0Alibaba1,075
16grok-imagine-image-qualityxAI1,045
17Qwen-Image-2.1Alibaba1,034
18MAI-Image-2.5-FlashMicrosoft1,033
19Qwen Image 2.0 Pro (2026-04-22)Alibaba1,030
20FLUX.2Black Forest Labs1,028

How this chart is built

Draws on a leading third-party evaluation programme that measures model capability across standardised test sets. The chart shows its latest published scores.

Positions are renumbered from high to low by this table's own metric, independent of any ranking the source publishes. Figures are the current real data with no manual adjustment; values we cannot obtain show as “—” and are never filled in with estimates.

FAQ

Artificial Analysis Image Models: who ranks #1?

In October 2026, OpenAI’s GPT-Image-2.5 Sunburst ranks #1 (score: 1,197); GPT-Image-2.5 Flare is second (1,190), followed by GPT Image 2 (1,172).

Artificial Analysis Image Models: what are the top 10?

Ranked by score as of October 2026, the top 10 are: 1. GPT-Image-2.5 Sunburst, 2. GPT-Image-2.5 Flare, 3. GPT Image 2, 4. Grok Imagine Image 2.0, 5. MAI-Image-2.6, 6. Nano Banana 2 (Gemini 3.1 Flash Image Preview), 7. Muse Image, 8. MAI-Image-2.6-Flash, 9. MAI-Image-2.5, 10. Nano Banana Pro (Gemini 3 Pro Image). The full top 20 is in the table on this page.

Artificial Analysis Image Models: where does the data come from, and how often is it updated?

Draws on a leading third-party evaluation programme that measures model capability across standardised test sets. The chart shows its latest published scores. Refreshed when the source publishes; the current data is dated October 1, 2026.

Related charts