AA-Image-Editing v2.1Artificial Analysis
Ranks image models on editing an image from a text instruction, by Elo from pairwise human preference votes.
New
6 models added in the last 30 days
Nano Banana 2.1, Ideogram 4.5 (High), HunyuanImage 3.0 Instruct, and 3 more
Range | Model | Elo | Samples | Released | API Pricing 1 | |
|---|---|---|---|---|---|---|
| 1 | 1 |  GPT Image 2.5 Sunburst (max) | 1198±1195% interval 1187 to 1209 | 17,534 | Sep 2026 | $210.7 /1k imgs |
| 2 | 2 | GPT Image 2.5 Flare (max) | 1178±1195% interval 1167 to 1189 | 18,372 | Sep 2026 | $210.7 /1k imgs |
| 3 | 3-5 |  MAI-Image-2.6 | 1148±1395% interval 1135 to 1161 | 15,808 | Aug 2026 | $38.9 /1k imgs |
| 4 | 3-5 |  Nano Banana 2.1 | 1147±1395% interval 1134 to 1160 | 6,356 | Oct 2026 | $33.6 /1k imgs |
| 5 | 3-6 | MAI-Image-2.6-Flash | 1141±1395% interval 1128 to 1154 | 10,166 | Sep 2026 | $19.5 /1k imgs |
| 6 | 5-9 | GPT Image 2 (high) | 1132±1395% interval 1119 to 1145 | 13,967 | Apr 2026 | $211.0 /1k imgs |
| 7 | 6-11 |  Muse Image | 1126±1295% interval 1114 to 1138 | 12,519 | Jul 2026 | $10.0 /1k imgs |
| 8 | 6-11 |  Seedream 5.0 Pro | 1126±1395% interval 1113 to 1139 | 12,798 | Jul 2026 | $90.0 /1k imgs |
| 9 | 6-12 | MAI-Image-2.5 | 1123±1395% interval 1110 to 1136 | 12,959 | Jun 2026 | $48.1 /1k imgs |
| 10 | 7-13 | Nano Banana 2 (Gemini 3.1 Flash Image) | 1120±1395% interval 1107 to 1133 | 15,818 | Feb 2026 | $67.0 /1k imgs |
| 11 | 8-13 | MAI-Image-2.5-Pro | 1117±1395% interval 1104 to 1130 | 14,814 | Jul 2026 | $108.5 /1k imgs |
| 12 | 9-14 | GPT Image 1.5 (high) | 1114±1395% interval 1101 to 1127 | 12,918 | Dec 2025 | $133.0 /1k imgs |
| 13 | 10-14 |  Grok Imagine Image 2.0 | 1111±1395% interval 1098 to 1124 | 7,852 | Aug 2026 | $60.0 /1k imgs |
| 14 | 12-15 | Nano Banana Pro (Gemini 3 Pro Image) | 1106±1395% interval 1093 to 1119 | 14,732 | Nov 2025 | $134.0 /1k imgs |
| 15 | 15-18 | MAI-Image-2.5-Flash | 1091±1395% interval 1078 to 1104 | 12,768 | Jun 2026 | $20.0 /1k imgs |
| 16 | 15-19 | grok-imagine-image-quality | 1090±1395% interval 1077 to 1103 | 8,054 | May 2026 | $50.0 /1k imgs |
| 17 | 15-19 |  HunyuanImage 3.5 (Preview) | 1089±1495% interval 1075 to 1103 | 5,823 | Sep 2026 | $24.0 /1k imgs |
| 18 | 15-20 |  Qwen-Image-3.0-Pro | 1084±1395% interval 1071 to 1097 | 12,906 | Jul 2026 | $40.0 /1k imgs |
| 19 | 16-22 | Qwen-Image-2.1  | 1079±1495% interval 1065 to 1093 | 5,844 | Sep 2026 | No API available |
| 20 | 18-23 |  Luma UNI 1 Max | 1075±1495% interval 1061 to 1089 | 5,925 | May 2026 | $100.0 /1k imgs |
| 21 | 19-25 | grok-imagine-image | 1069±1495% interval 1055 to 1083 | 5,522 | Jan 2026 | $20.0 /1k imgs |
| 22 | 20-27 |  Ideogram 4.5 (High) | 1064±1695% interval 1048 to 1080 | 2,781 | Sep 2026 | $220.0 /1k imgs |
| 23 | 21-27 | Qwen-Image-3.0 | 1062±1395% interval 1049 to 1075 | 7,526 | Jul 2026 | $30.0 /1k imgs |
| 24 | 21-27 | Seedream 5.0 Lite | 1061±1495% interval 1047 to 1075 | 5,896 | Feb 2026 | $35.0 /1k imgs |
| 25 | 21-27 | HunyuanImage 3.0 Instruct | 1060±1495% interval 1046 to 1074 | 5,519 | Jan 2026 | No API available |
| 26 | 20-30 | Qwen Image 2.0 Pro (2026-04-22) | 1058±1895% interval 1040 to 1076 | 2,350 | Apr 2026 | $75.0 /1k imgs |
| 27 | 25-32 | Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | 1048±1395% interval 1035 to 1061 | 7,311 | Jun 2026 | $33.6 /1k imgs |
| 28 | 26-33 | Wan 2.7 Pro | 1045±1495% interval 1031 to 1059 | 5,749 | Apr 2026 | $75.0 /1k imgs |
| 29 | 26-33 | Wan 2.6 Image | 1044±1495% interval 1030 to 1058 | 5,709 | Dec 2025 | $30.0 /1k imgs |
| 30 | 26-33 |  DreamImage 2.0 | 1044±1495% interval 1030 to 1058 | 5,677 | Jul 2026 | $80.0 /1k imgs |
| 31 | 27-33 | Luma UNI 1 | 1042±1395% interval 1029 to 1055 | 8,923 | May 2026 | $40.4 /1k imgs |
| 32 | 27-34 | Seedream 4.5 | 1040±1495% interval 1026 to 1054 | 4,923 | Dec 2025 | $40.0 /1k imgs |
| 33 | 28-34 | Wan 2.7 | 1038±1395% interval 1025 to 1051 | 7,074 | Apr 2026 | $26.0 /1k imgs |
| 34 | 32-38 | HunyuanImage 3.0 (Tencent Cloud) | 1023±1795% interval 1006 to 1040 | 2,530 | Sep 2025 | $32.0 /1k imgs |
| 35 | 32-40 |  FLUX.2 [dev] Flash | 1020±2095% interval 1000 to 1040 | 2,007 | Dec 2025 | $5.0 /1k imgs |
| 36 | 34-42 | FLUX.2 [dev] Turbo | 1013±2095% interval 993 to 1033 | 2,036 | Dec 2025 | $8.0 /1k imgs |
| 37 | 34-40 |  FLUX.2 [klein] 9B | 1012±1395% interval 999 to 1025 | 6,464 | Jan 2026 | $17.0 /1k imgs |
| 38 | 35-43 | FLUX.2 [pro] | 1006±1795% interval 989 to 1023 | 3,516 | Nov 2025 | $45.0 /1k imgs |
| 39 | 35-42 |  Step Image Edit 2 | 1006±1495% interval 992 to 1020 | 4,971 | May 2026 | $3.0 /1k imgs |
| 40 | 35-45 | FLUX.2 [flex] | 1001±1995% interval 982 to 1020 | 2,188 | Nov 2025 | $120.0 /1k imgs |
| 41 | 37-44 | FLUX.2 [dev] | 1000AnchorFixed at 1000 as the reference point for this board's Elo scale. | 8,333 | Nov 2025 | $24.0 /1k imgs |
| 42 | 39-45 | FLUX.2 [max] | 995±1395% interval 982 to 1008 | 6,296 | Dec 2025 | $140.0 /1k imgs |
| 43 | 38-48 | GPT Image 1 (high) | 990±1995% interval 971 to 1009 | 2,141 | Apr 2025 | $167.0 /1k imgs |
| 44 | 40-49 | Qwen Image Edit Plus 2511 | 985±2095% interval 965 to 1005 | 2,181 | Nov 2025 | $60.0 /1k imgs |
| 45 | 40-50 | Qwen Image Edit Max 2601 | 983±1995% interval 964 to 1002 | 2,229 | Jan 2026 | $75.0 /1k imgs |
| 46 | 42-49 | Nano Banana (Gemini 2.5 Flash Image) | 983±1495% interval 969 to 997 | 4,684 | Aug 2025 | $39.0 /1k imgs |
| 47 | 42-49 |  P-Image-Edit | 982±1495% interval 968 to 996 | 4,631 | Nov 2025 | $10.0 /1k imgs |
| 48 | 42-50 |  Reve V1 (December) | 980±2095% interval 960 to 1000 | 2,239 | Dec 2025 | $40.0 /1k imgs |
| 49 | 45-52 | Qwen Image 2.0 (2026-03-03) | 968±2095% interval 948 to 988 | 2,273 | Mar 2026 | $35.0 /1k imgs |
| 50 | 45-53 | Qwen Image Edit Plus 2509 | 966±2095% interval 946 to 986 | 2,031 | Sep 2025 | $30.0 /1k imgs |
| 51 | 50-54 |  HiDream-O1-Image | 952±1595% interval 937 to 967 | 4,675 | May 2026 | $10.0 /1k imgs |
| 52 | 50-54 |  Kling Image 3.0 Omni | 951±1595% interval 936 to 966 | 4,637 | Feb 2026 | $28.0 /1k imgs |
| 53 | 51-55 | Kling Image O1 | 945±1595% interval 930 to 960 | 4,561 | Dec 2025 | $28.0 /1k imgs |
| 54 | 50-57 |  Vidu Q2 | 941±2095% interval 921 to 961 | 2,107 | Nov 2025 | $40.0 /1k imgs |
| 55 | 52-58 | FLUX.2 [klein] 4B | 933±1895% interval 915 to 951 | 3,065 | Jan 2026 | $15.0 /1k imgs |
| 56 | 54-61 | GPT Image 1 Mini (medium) | 922±2295% interval 900 to 944 | 1,893 | Oct 2025 | $11.0 /1k imgs |
| 57 | 54-60 | FLUX.1 Kontext [max] | 922±2195% interval 901 to 943 | 2,081 | May 2025 | $80.0 /1k imgs |
| 58 | 55-62 |  FIBO Edit | 914±2295% interval 892 to 936 | 2,038 | Jan 2026 | $40.0 /1k imgs |
| 59 | 55-62 | FLUX.1 Kontext [pro] | 913±2295% interval 891 to 935 | 1,978 | May 2025 | $40.0 /1k imgs |
| 60 | 57-63 | SeedEdit 3.0 | 903±2295% interval 881 to 925 | 1,978 | Jun 2025 | $0.0 /1k imgs |
| 61 | 58-64 |  LongCat Image | 893±2295% interval 871 to 915 | 1,962 | Dec 2025 | $130.0 /1k imgs |
| 62 | 59-64 | Qwen Image Edit | 893±2295% interval 871 to 915 | 2,068 | Aug 2025 | $30.0 /1k imgs |
| 63 | 59-64 | FLUX.1 Kontext [dev] | 891±2395% interval 868 to 914 | 1,845 | Jun 2025 | $25.0 /1k imgs |
| 64 | 62-64 | HiDream-O1-Image-Dev | 877±1595% interval 862 to 892 | 4,650 | May 2026 | $5.0 /1k imgs |
| 65 | 65-66 |  GLM-Image | 818±2595% interval 793 to 843 | 1,972 | Jan 2026 | $50.0 /1k imgs |
| 66 | 65-68 | Bagel | 790±2895% interval 762 to 818 | 1,856 | May 2025 | $100.0 /1k imgs |
| 67 | 65-68 |  OmniGen V2 | 788±2695% interval 762 to 814 | 2,073 | Jun 2025 | $150.0 /1k imgs |
| 68 | 66-68 | HiDream-E1.1 | 777±2995% interval 748 to 806 | 1,781 | Jul 2025 | $60.0 /1k imgs |
| 69 | 69 | Step1X-Edit | 603±4395% interval 560 to 646 | 1,946 | Apr 2025 | $30.0 /1k imgs |
1 API Pricing reflects the cost to generate 1,000 images on the model creator's API at 1024x1024 with the model's default settings
Frequently Asked Questions
The top models on AA-Image-Editing v2.1 by Elo rating are: 1. GPT Image 2.5 Sunburst (max) (Elo 1198), 2. GPT Image 2.5 Flare (max) (Elo 1178), 3. MAI-Image-2.6 (Elo 1148), 4. Nano Banana 2.1 (Elo 1147), and 5. MAI-Image-2.6-Flash (Elo 1141). Rankings are based on blind votes from our validated private evaluators.
Models are ranked using an Elo rating system derived from blind comparisons judged by our validated private evaluators. Voters compare edited outputs from the same input image and editing instruction, choosing the result they prefer. Higher Elo scores indicate a model is preferred more often.