Best AI Image Generators: Text-to-Image Leaderboard

Ranked by LMArena scores from anonymous head-to-head human votes. Artificial Analysis Image Arena Elo is shown for comparison; the two scales use different baselines and cannot be compared directly.

Text-to-image Image editing
🏆 #1 Image editingData Oct 20, 2026
GPT-Image-2.5 SunburstOpenAI · 1514

On LMArena’s text-to-image leaderboard published Oct 7, 2026, #1 is GPT Image 2.5 Sunburst by OpenAI (1425, 17,752 votes); the image-editing leader is GPT-Image-2.5 Sunburst (1514).

Full leaderboard · Image editing

60 models
All Open weights

Sorted by LMArena score by default; ± is half the 95% confidence interval.

# Model LMArena Artificial Analysis Released License
Score Rank Votes Elo Rank
1 GPT-Image-2.5 SunburstOpenAI 1514±6 1 82,078 1199±11 1 Sep 2026 Proprietary
2 GPT-Image-2.5 FlareOpenAI 1481±5 2 79,506 1182±11 2 Sep 2026 Proprietary
3 GPT-Image-2 (Medium)OpenAI 1461±3 3 328,307 1136±13 7 Apr 2026 Proprietary
4 Grok Imagine Image 2.0xAI 1439±6 4 30,542 – – – Proprietary
5 Gemini Nano Banana 2.1Google 1426±6 5 12,938 1151±13 5 Oct 2026 Proprietary
6 MAI-Image-2.6Microsoft AI 1425±5 6 69,611 1151±12 3 Aug 2026 Proprietary
7 Grok Imagine Image 2.0 (20260801)xAI 1415±5 7 76,730 – – – Proprietary
8 Krea 3 UltraKrea 1411±11 8 2,619 – – – –
9 MAI-Image-2.5Microsoft AI 1400±4 9 190,119 1127±13 10 Jun 2026 Proprietary
10 Muse ImageMeta 1398±4 10 164,982 1129±12 9 Jul 2026 Proprietary
11 Seedream 5.0 ProByteDance 1391±4 11 307,763 1131±13 8 Jul 2026 Proprietary
12 Grok Imagine Image (Quality, 20260519)xAI 1391±6 12 38,207 – – – Proprietary
13 Gemini 3 Pro Image (nano-banana-pro, 2K)Google 1390±3 13 676,298 – – – Proprietary
14 Gemini 3.1 Flash Image (nano-banana-2, Web Search)Google 1387±3 14 253,535 – – – Proprietary
15 ChatGPT Image Latest (High Fidelity, 20251216)OpenAI 1387±3 15 651,019 – – – Proprietary
16 Gemini 3 Pro Image Preview (nano-banana-pro)Google 1385±3 16 540,728 – – – Proprietary
17 Reve 2.1Reve 1373±6 17 20,810 – – – Proprietary
18 GPT-Image-1.5 (High Fidelity)OpenAI 1367±3 18 676,312 1119±13 15 Dec 2025 Proprietary
19 Qwen-Image-2.1AlibabaOpen 1362±6 19 36,713 1081±13 23 Sep 2026 qwen-research
20 Reve 2.0Reve 1358±6 20 19,270 – – – Proprietary
21 Ideogram 4.5Ideogram 1347±5 21 47,127 1070±15 27 Sep 2026 Proprietary
22 UNI-1.1 MaxLuma AI 1334±5 22 43,459 – – – Proprietary
23 Grok Imagine ImagexAI 1329±3 23 792,122 – – – Proprietary
24 UNI-1.1Luma AI 1316±3 24 210,418 – – – Proprietary
25 Gemini 3.1 Flash-Lite Image (nano-banana-2-lite)Google 1313±5 25 65,657 – – – Proprietary
26 Seedream 4.5ByteDance 1303±2 26 1,342,849 1043±13 38 Dec 2025 Proprietary
27 Qwen-Image-2.0-Pro (2026-06-22)Alibaba 1302±5 27 56,713 1058±17 33 Apr 2026 Proprietary
28 Wan 2.7 Image ProAlibaba 1302±4 28 43,966 – – – Proprietary
29 HunyuanImage 3.0 InstructTencentOpen 1302±3 29 331,438 1063±13 31 Jan 2026 tencent-hunyuan-community
30 Wan 2.7 ImageAlibaba 1301±4 30 44,788 – – – Proprietary
31 Seedream 5.0 LiteByteDance 1295±3 31 485,179 1062±13 32 Feb 2026 Proprietary
32 Gemini 2.5 Flash Image Preview (nano-banana)Google 1291±2 32 11,430,191 – – – Proprietary
33 Seedream 4.0 (2K)ByteDance 1267±7 33 210,030 – – – Proprietary
34 FLUX.2 [max]Black Forest Labs 1261±3 34 375,802 – – – Proprietary
35 Reve v1.1Reve 1261±3 35 672,619 – – – Proprietary
36 Kling Image O1Kling (Kuaishou) 1250±4 36 143,646 944±14 61 Dec 2025 Proprietary
37 FLUX.2 [pro]Black Forest Labs 1247±3 37 663,378 – – – Proprietary
38 Qwen-Image-EditAlibabaOpen 1241±3 38 2,021,780 – – – Apache 2.0
39 Reve v1Reve 1235±5 39 395,141 987±18 52 Dec 2025 Proprietary
40 Qwen-Image-Edit (2511)AlibabaOpen 1233±3 40 485,092 – – – Apache 2.0
41 Wan 2.6 ImageAlibaba 1231±3 41 703,228 1044±13 37 Dec 2025 Proprietary
42 FLUX.2 [flex]Black Forest Labs 1224±3 42 443,870 – – – Proprietary
43 FLUX.2 [klein] 9BBlack Forest LabsOpen 1224±3 43 540,679 1015±13 43 Jan 2026 flux-non-commercial-license
44 FLUX.2 [dev]Black Forest LabsOpen 1224±4 44 254,229 – – – flux-non-commercial-license
45 Seedream 4.0 (High Res, fal)ByteDance 1216±3 45 1,229,694 – – – Proprietary
46 Seedream 4.0 (fal)ByteDance 1210±6 46 152,100 – – – Proprietary
47 ?P-Image-Edit 1210±3 47 359,855 985±14 53 Nov 2025 Proprietary
48 Reve v1.1 FastReve 1207±3 48 553,281 – – – Proprietary
49 Reve Edit FastReve 1198±4 49 230,147 – – – Proprietary
50 FLUX.2 [klein] 4BBlack Forest LabsOpen 1187±3 50 541,237 938±17 62 Jan 2026 Apache 2.0
51 FLUX.1 Kontext [max]Black Forest Labs 1182±3 51 390,852 – – – Proprietary
52 Wan 2.5 PreviewAlibaba 1180±3 52 723,477 – – – Proprietary
53 FLUX.1 Kontext [pro]Black Forest Labs 1177±3 53 6,404,500 – – – Proprietary
54 FLUX.1 Kontext [dev]Black Forest LabsOpen 1151±3 54 3,639,665 – – – flux-1-dev-non-commercial-license
55 SeedEdit 3.0ByteDance 1139±3 55 4,918,211 901±20 67 Jun 2025 Proprietary
56 GPT-Image-1OpenAI 1139±3 56 2,869,690 995±18 50 Apr 2025 Proprietary
57 GPT-Image-1 miniOpenAI 1124±3 57 682,782 934±19 63 Oct 2025 Proprietary
58 Gemini 2.0 Flash Preview Image GenerationGoogle 1082±2 58 4,942,749 – – – Proprietary
59 BAGELByteDanceOpen 1027±6 59 13,488 792±25 73 May 2025 Apache 2.0
60 Step1X-EditStepFunOpen 996±5 60 155,416 601±39 76 Apr 2025 Apache 2.0

Scores: LMArena leaderboard-dataset (lmarena-ai on Hugging Face, CC-BY-4.0), Image editing published Oct 20, 2026; AA columns from the Artificial Analysis Image Arena. Please credit the sources when citing.

Sources & methodology

  1. Scores come from LMArena’s text_to_image and image_edit leaderboards (overall category, latest release): users vote between two anonymous outputs and LMArena fits a Bradley–Terry model.
  2. AA values are shown only when the model name (ignoring bracketed variant notes) maps to exactly one row; hover to see the AA entry name. Anything unmatched stays blank (–); nothing is guessed.
  3. Price, resolution and generation speed have no reliable public source yet and are not shown.
  4. Synced daily at 06:55 UTC; public JSON: /api/public/image-leaderboard.php

FAQ

What is the best AI image generator right now?

On LMArena’s text-to-image leaderboard published Oct 7, 2026, #1 is GPT Image 2.5 Sunburst by OpenAI (1425, 17,752 votes); the image-editing leader is GPT-Image-2.5 Sunburst (1514).

Which vendors rank highest for image generation?

On the text-to-image board, each vendor's best model ranks: OpenAI GPT Image 2.5 Sunburst (1425, #1), xAI Grok Imagine Image 2.0 (Canvas) (1334, #4), Microsoft AI MAI Image 2.6 (1332, #5), Google Gemini Nano Banana 2.1 (1328, #6), Reve Reve 2.1 (1302, #7).

Which open-weight image models can I run locally?

19 text-to-image models have non-proprietary licenses; the highest is Qwen Image 2.1 (1223, license qwen-research). Some licenses (e.g. FLUX dev non-commercial) restrict commercial use.

Why do LMArena and AA scores differ so much?

The two platforms use different voters, prompts and baselines, so absolute numbers are not comparable; relative order within one platform is what matters.