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
8 models added in the last 30 days
Qwen-Image-2.1-Pro, Qwen-Image-2.1-Turbo, Nano Banana 2.1, and 5 more
Range | Model | Elo | Samples | Released | API Pricing 1 | |
|---|---|---|---|---|---|---|
| 1 | 1 |  GPT Image 2.5 Sunburst (max) | 1199±1195% interval 1188 to 1210 | 18,115 | Sep 2026 | $210.7 /1k imgs |
| 2 | 2 | GPT Image 2.5 Flare (max) | 1182±1195% interval 1171 to 1193 | 19,014 | Sep 2026 | $210.7 /1k imgs |
| 3 | 3-5 |  Nano Banana 2.1 | 1152±1395% interval 1139 to 1165 | 6,745 | Oct 2026 | $33.6 /1k imgs |
| 4 | 3-5 |  MAI-Image-2.6 | 1152±1295% interval 1140 to 1164 | 16,528 | Aug 2026 | $38.9 /1k imgs |
| 5 | 3-7 | MAI-Image-2.6-Flash | 1144±1395% interval 1131 to 1157 | 10,638 | Sep 2026 | $19.5 /1k imgs |
| 6 | 5-9 | GPT Image 2 (high) | 1138±1395% interval 1125 to 1151 | 14,435 | Apr 2026 | $211.0 /1k imgs |
| 7 | 5-11 |  Seedream 5.0 Pro | 1133±1395% interval 1120 to 1146 | 13,244 | Jul 2026 | $90.0 /1k imgs |
| 8 | 6-11 |  Muse Image | 1131±1295% interval 1119 to 1143 | 13,227 | Jul 2026 | $10.0 /1k imgs |
| 9 | 6-12 | MAI-Image-2.5 | 1128±1395% interval 1115 to 1141 | 13,416 | Jun 2026 | $48.1 /1k imgs |
| 10 | 7-13 | Nano Banana 2 (Gemini 3.1 Flash Image) | 1124±1295% interval 1112 to 1136 | 16,514 | Feb 2026 | $67.0 /1k imgs |
| 11 | 8-13 | MAI-Image-2.5-Pro | 1123±1295% interval 1111 to 1135 | 15,510 | Jul 2026 | $108.5 /1k imgs |
| 12 | 9-14 | GPT Image 1.5 (high) | 1119±1395% interval 1106 to 1132 | 13,388 | Dec 2025 | $133.0 /1k imgs |
| 13 | 9-14 |  Grok Imagine Image 2.0 | 1118±1395% interval 1105 to 1131 | 8,561 | Aug 2026 | $60.0 /1k imgs |
| 14 | 13-15 | Nano Banana Pro (Gemini 3 Pro Image) | 1107±1295% interval 1095 to 1119 | 15,429 | Nov 2025 | $134.0 /1k imgs |
| 15 | 14-18 | grok-imagine-image-quality | 1096±1395% interval 1083 to 1109 | 8,656 | May 2026 | $50.0 /1k imgs |
| 16 | 15-19 | MAI-Image-2.5-Flash | 1093±1395% interval 1080 to 1106 | 13,238 | Jun 2026 | $20.0 /1k imgs |
| 17 | 15-19 |  HunyuanImage 3.5 (Preview) | 1093±1395% interval 1080 to 1106 | 6,272 | Sep 2026 | $24.0 /1k imgs |
| 18 | 15-20 |  Qwen-Image-3.0-Pro | 1091±1395% interval 1078 to 1104 | 13,389 | Jul 2026 | $40.0 /1k imgs |
| 19 | 16-22 | Qwen-Image-2.1  | 1083±1395% interval 1070 to 1096 | 6,312 | Sep 2026 | No API available |
| 20 | 17-23 |  Luma UNI 1 Max | 1080±1395% interval 1067 to 1093 | 6,364 | May 2026 | $100.0 /1k imgs |
| 21 | 19-26 | grok-imagine-image | 1074±1395% interval 1061 to 1087 | 5,968 | Jan 2026 | $20.0 /1k imgs |
| 22 | 20-27 | Qwen-Image-2.1-Pro | 1070±1395% interval 1057 to 1083 | 5,255 | Oct 2026 | $35.3 /1k imgs |
| 23 | 19-28 |  Ideogram 4.5 (High) | 1070±1595% interval 1055 to 1085 | 3,205 | Sep 2026 | $220.0 /1k imgs |
| 24 | 20-28 | Qwen-Image-3.0 | 1069±1395% interval 1056 to 1082 | 7,952 | Jul 2026 | $30.0 /1k imgs |
| 25 | 21-29 | Qwen-Image-2.1-Turbo | 1065±1495% interval 1051 to 1079 | 4,895 | Oct 2026 | $16.0 /1k imgs |
| 26 | 22-29 | HunyuanImage 3.0 Instruct | 1064±1395% interval 1051 to 1077 | 5,986 | Jan 2026 | No API available |
| 27 | 22-30 | Seedream 5.0 Lite | 1062±1395% interval 1049 to 1075 | 6,350 | Feb 2026 | $35.0 /1k imgs |
| 28 | 22-33 | Qwen Image 2.0 Pro (2026-04-22) | 1058±1795% interval 1041 to 1075 | 2,773 | Apr 2026 | $75.0 /1k imgs |
| 29 | 26-33 | Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | 1055±1395% interval 1042 to 1068 | 7,962 | Jun 2026 | $33.6 /1k imgs |
| 30 | 27-35 | Wan 2.7 Pro | 1049±1395% interval 1036 to 1062 | 6,192 | Apr 2026 | $75.0 /1k imgs |
| 31 | 28-35 |  DreamImage 2.0 | 1048±1395% interval 1035 to 1061 | 6,140 | Jul 2026 | $80.0 /1k imgs |
| 32 | 29-36 | Wan 2.6 Image | 1045±1395% interval 1032 to 1058 | 6,125 | Dec 2025 | $30.0 /1k imgs |
| 33 | 29-36 | Seedream 4.5 | 1045±1495% interval 1031 to 1059 | 5,352 | Dec 2025 | $40.0 /1k imgs |
| 34 | 30-36 | Luma UNI 1 | 1043±1395% interval 1030 to 1056 | 9,338 | May 2026 | $40.4 /1k imgs |
| 35 | 30-36 | Wan 2.7 | 1043±1395% interval 1030 to 1056 | 7,493 | Apr 2026 | $26.0 /1k imgs |
| 36 | 32-38 | HunyuanImage 3.0 (Tencent Cloud) | 1032±1695% interval 1016 to 1048 | 2,966 | Sep 2025 | $32.0 /1k imgs |
| 37 | 36-42 |  FLUX.2 [dev] Turbo | 1019±1895% interval 1001 to 1037 | 2,425 | Dec 2025 | $8.0 /1k imgs |
| 38 | 37-42 |  FLUX.2 [klein] 9B | 1015±1395% interval 1002 to 1028 | 7,049 | Jan 2026 | $17.0 /1k imgs |
| 39 | 36-44 | FLUX.2 [dev] Flash | 1013±1895% interval 995 to 1031 | 2,451 | Dec 2025 | $5.0 /1k imgs |
| 40 | 37-43 | FLUX.2 [pro] | 1013±1595% interval 998 to 1028 | 4,096 | Nov 2025 | $45.0 /1k imgs |
| 41 | 37-44 |  Step Image Edit 2 | 1009±1495% interval 995 to 1023 | 5,376 | May 2026 | $3.0 /1k imgs |
| 42 | 40-46 | FLUX.2 [max] | 1001±1395% interval 988 to 1014 | 6,819 | Dec 2025 | $140.0 /1k imgs |
| 43 | 40-47 | FLUX.2 [dev] | 1000AnchorFixed at 1000 as the reference point for this board's Elo scale. | 8,882 | Nov 2025 | $24.0 /1k imgs |
| 44 | 39-49 | FLUX.2 [flex] | 999±1895% interval 981 to 1017 | 2,591 | Nov 2025 | $120.0 /1k imgs |
| 45 | 39-49 | GPT Image 1 (high) | 997±1895% interval 979 to 1015 | 2,506 | Apr 2025 | $167.0 /1k imgs |
| 46 | 41-50 | Qwen Image Edit Plus 2511 | 990±1895% interval 972 to 1008 | 2,600 | Nov 2025 | $60.0 /1k imgs |
| 47 | 43-51 |  Reve V1 (December) | 987±1895% interval 969 to 1005 | 2,646 | Dec 2025 | $40.0 /1k imgs |
| 48 | 44-51 |  P-Image-Edit | 986±1495% interval 972 to 1000 | 5,029 | Nov 2025 | $10.0 /1k imgs |
| 49 | 45-51 | Nano Banana (Gemini 2.5 Flash Image) | 984±1495% interval 970 to 998 | 5,072 | Aug 2025 | $39.0 /1k imgs |
| 50 | 44-52 | Qwen Image Edit Max 2601 | 982±1895% interval 964 to 1000 | 2,658 | Jan 2026 | $75.0 /1k imgs |
| 51 | 48-54 | Qwen Image 2.0 (2026-03-03) | 969±1895% interval 951 to 987 | 2,652 | Mar 2026 | $35.0 /1k imgs |
| 52 | 49-55 | Qwen Image Edit Plus 2509 | 965±1995% interval 946 to 984 | 2,453 | Sep 2025 | $30.0 /1k imgs |
| 53 | 51-56 |  HiDream-O1-Image | 956±1495% interval 942 to 970 | 5,074 | May 2026 | $10.0 /1k imgs |
| 54 | 52-57 |  Kling Image 3.0 Omni | 953±1495% interval 939 to 967 | 5,057 | Feb 2026 | $28.0 /1k imgs |
| 55 | 51-58 |  Vidu Q2 | 953±1995% interval 934 to 972 | 2,481 | Nov 2025 | $40.0 /1k imgs |
| 56 | 53-58 | Kling Image O1 | 946±1495% interval 932 to 960 | 4,965 | Dec 2025 | $28.0 /1k imgs |
| 57 | 54-59 | FLUX.2 [klein] 4B | 940±1795% interval 923 to 957 | 3,510 | Jan 2026 | $15.0 /1k imgs |
| 58 | 54-60 | GPT Image 1 Mini (medium) | 937±2095% interval 917 to 957 | 2,279 | Oct 2025 | $11.0 /1k imgs |
| 59 | 56-61 | FLUX.1 Kontext [max] | 928±1995% interval 909 to 947 | 2,490 | May 2025 | $80.0 /1k imgs |
| 60 | 58-63 | FLUX.1 Kontext [pro] | 916±2095% interval 896 to 936 | 2,381 | May 2025 | $40.0 /1k imgs |
| 61 | 58-63 |  FIBO Edit | 915±2095% interval 895 to 935 | 2,460 | Jan 2026 | $40.0 /1k imgs |
| 62 | 60-65 | SeedEdit 3.0 | 903±2095% interval 883 to 923 | 2,368 | Jun 2025 | $0.0 /1k imgs |
| 63 | 60-65 |  LongCat Image | 899±2095% interval 879 to 919 | 2,354 | Dec 2025 | $130.0 /1k imgs |
| 64 | 62-66 | Qwen Image Edit | 889±2095% interval 869 to 909 | 2,452 | Aug 2025 | $30.0 /1k imgs |
| 65 | 62-66 | FLUX.1 Kontext [dev] | 884±2195% interval 863 to 905 | 2,223 | Jun 2025 | $25.0 /1k imgs |
| 66 | 64-66 | HiDream-O1-Image-Dev | 881±1595% interval 866 to 896 | 5,039 | May 2026 | $5.0 /1k imgs |
| 67 | 67-68 |  GLM-Image | 824±2395% interval 801 to 847 | 2,383 | Jan 2026 | $50.0 /1k imgs |
| 68 | 67-70 | Bagel | 793±2595% interval 768 to 818 | 2,221 | May 2025 | $100.0 /1k imgs |
| 69 | 68-70 |  OmniGen V2 | 787±2495% interval 763 to 811 | 2,463 | Jun 2025 | $150.0 /1k imgs |
| 70 | 68-70 | HiDream-E1.1 | 772±2695% interval 746 to 798 | 2,152 | Jul 2025 | $60.0 /1k imgs |
| 71 | 71 | Step1X-Edit | 600±4095% interval 560 to 640 | 2,351 | 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 1199), 2. GPT Image 2.5 Flare (max) (Elo 1182), 3. Nano Banana 2.1 (Elo 1152), 4. MAI-Image-2.6 (Elo 1152), and 5. MAI-Image-2.6-Flash (Elo 1144). 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.