Benchmark Almanac

DesignArena Graphic Design

GPT-Image-2.5 Sunburst 1,436
Not in model cards

Image models make a graphic design piece from a user prompt. Users compare two anonymous designs and vote for the better one.

One of 40 DesignArena boards. DesignArena ranks entries with Bradley-Terry ratings from user votes.

Kind
Image generation and editing
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Scores refreshed
November 12th, 2026
78
Entries on its own leaderboard, 43 models we track

GPT-Image-2.5 Sunburst leads with 1,436

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

78 entries. Scroll sideways for the rest. Bars start at 800.

  1. 1,436 GPT-Image-2.5 Sunburst
  2. 1,435 GPT-Image-2
  3. 1,413 GPT-Image-2.5 Flare
  4. 1,411 Riverflow 2.5 Pro
  5. 1,372 Reve 2.1
  6. 1,352 Reve 2.0
  7. 1,337 Grok Imagine Image 2.0
  8. 1,333 Nano Banana 2.1
  9. 1,330 MAI-Image-2.6
  10. 1,314 Grok Imagine Image 2 Auto
  11. 1,308 MAI-Image-2.6 Flash
  12. 1,295 GPT-Image-1.5
  13. 1,279 Muse Image
  14. 1,276 Seedream 5.0 Pro
  15. 1,273 Qwen-Image-3.0-Pro
  16. 1,261 Gemini 3.1 Flash Image Gen 2K (Nano Banana 2)
  17. 1,257 Ideogram 4.0
  18. 1,254 FLUX 3 Image
  19. 1,251 Gemini 3 Pro Image Gen 2K (Nano Banana Pro)
  20. 1,249 Gemini 3.1 Flash-Lite Image (nano-banana-2-lite)
  21. 1,249 Gemini 3.1 Flash Image Gen (Nano Banana 2)
  22. 1,244 UNI-1.1 Max
  23. 1,242 UNI-1.1
  24. 1,226 Gemini 3 Pro Image Preview
  25. 1,210 MAI-Image-2.5
  26. 1,200 Recraft V4.1 Utility Pro
  27. 1,190 Seedream Lite 5.0
  28. 1,189 FLUX.2 [flex]
  29. 1,174 Imagen 4 Ultra Generate Preview 06-06
  30. 1,173 Recraft V4 Pro
  31. 1,172 FLUX.2 [pro]
  32. 1,172 GPT-Image-1
  33. 1,171 Krea 2 Large
  34. 1,167 Gemini 2.5 Flash Image Gen (Nano Banana)
  35. 1,165 GPT-Image-1 mini
  36. 1,162 Grok Imagine
  37. 1,159 Imagen 4 Ultra (generate-001)
  38. 1,157 ImagineArt 2.0
  39. 1,156 Krea 2 Medium
  40. 1,152 Recraft V4.1 Pro
  41. 1,151 Gemini 2.5 Flash Image (Nano Banana)
  42. 1,150 Imagen 4 Generate Preview 06-06
  43. 1,148 Seedream 4.0
  44. 1,146 p-image-ideogram
  45. 1,145 Wan 2.7 Pro
  46. 1,143 Wan 2.7
  47. 1,138 Krea 2 Turbo
  48. 1,137 Imagen 4 (generate-001)
  49. 1,130 Recraft V4.1
  50. 1,128 FLUX.1 Kontext [max]
  51. 1,109 GLM-Image
  52. 1,096 Recraft V3
  53. 1,089 Reve Create
  54. 1,086 FLUX.1 Kontext [pro]
  55. 1,086 FLUX.2 [dev]
  56. 1,085 HunyuanImage 3.0
  57. 1,084 Recraft V4.1 Flash
  58. 1,080 ERNIE Image Turbo
  59. 1,077 ERNIE Image
  60. 1,076 Boogu-Image-0.1
  61. 1,072 Qwen-Image (2512)
  62. 1,065 Cosmos 3 Super
  63. 1,049 DALL-E 3
  64. 1,043 Vidu Q2
  65. 1,036 FLUX.2 [klein] 9B
  66. 1,028 Z-Image-Turbo
  67. 1,022 Ideogram 3.0
  68. 1,017 Nucleus Image
  69. 1,013 FLUX.1 [pro] Ultra
  70. 1,005 FLUX.2 [klein] 4B
  71. 1,000 FLUX.2 Klein 4B Distilled
  72. 1,000 Grok 2 Image Gen
  73. 990 FLUX.1 Krea Dev
  74. 963 Imagen 4 Fast Generate 001
  75. 962 P-Image
  76. 953 Qwen-Image
  77. 937 HiDream-O1-Image
  78. 903 HiDream O1 Image Dev

Bars start at 800.

  1. 1 GPT-Image-2.5 Sunburst 1,436
  2. 2 GPT-Image-2 1,435
  3. 3 GPT-Image-2.5 Flare 1,413
  4. 4 Riverflow 2.5 Pro 1,411
  5. 5 Reve 2.1 1,372
  6. 6 Reve 2.0 1,352
  7. 7 Grok Imagine Image 2.0 medium 1,337
  8. 8 Nano Banana 2.1 1,333
  9. 9 MAI-Image-2.6 1,330
  10. 10 Grok Imagine Image 2 Auto 1,314
  11. 11 MAI-Image-2.6 Flash 1,308
  12. 12 GPT-Image-1.5 1,295
  13. 13 Muse Image 1,279
  14. 14 Seedream 5.0 Pro 1,276
  15. 15 Qwen-Image-3.0-Pro 1,273
  16. 16 Gemini 3.1 Flash Image Gen 2K (Nano Banana 2) 1,261
  17. 17 Ideogram 4.0 1,257
  18. 18 FLUX 3 Image 1,254
  19. 19 Gemini 3 Pro Image Gen 2K (Nano Banana Pro) 1,251
  20. 20 Gemini 3.1 Flash-Lite Image (nano-banana-2-lite) 1,249
  21. 21 Gemini 3.1 Flash Image Gen (Nano Banana 2) 1,249
  22. 22 UNI-1.1 Max 1,244
  23. 23 UNI-1.1 1,242
  24. 24 Gemini 3 Pro Image Preview 1,226
  25. 25 MAI-Image-2.5 1,210
  26. 26 Recraft V4.1 Utility Pro 1,200
  27. 27 Seedream Lite 5.0 1,190
  28. 28 FLUX.2 [flex] 1,189
  29. 29 Imagen 4 Ultra Generate Preview 06-06 1,174
  30. 30 Recraft V4 Pro 1,173
  31. 31 FLUX.2 [pro] 1,172
  32. 32 GPT-Image-1 1,172
  33. 33 Krea 2 Large 1,171
  34. 34 Gemini 2.5 Flash Image Gen (Nano Banana) 1,167
  35. 35 GPT-Image-1 mini 1,165
  36. 36 Grok Imagine 1,162
  37. 37 Imagen 4 Ultra (generate-001) 1,159
  38. 38 ImagineArt 2.0 1,157
  39. 39 Krea 2 Medium 1,156
  40. 40 Recraft V4.1 Pro 1,152
  41. 41 Gemini 2.5 Flash Image (Nano Banana) 1,151
  42. 42 Imagen 4 Generate Preview 06-06 1,150
  43. 43 Seedream 4.0 1,148
  44. 44 p-image-ideogram medium, high 1,146
  45. 45 Wan 2.7 Pro 1,145
  46. 46 Wan 2.7 1,143
  47. 47 Krea 2 Turbo 1,138
  48. 48 Imagen 4 (generate-001) 1,137
  49. 49 Recraft V4.1 1,130
  50. 50 FLUX.1 Kontext [max] 1,128
  51. 51 GLM-Image 1,109
  52. 52 Recraft V3 1,096
  53. 53 Reve Create 1,089
  54. 54 FLUX.1 Kontext [pro] 1,086
  55. 55 FLUX.2 [dev] 1,086
  56. 56 HunyuanImage 3.0 1,085
  57. 57 Recraft V4.1 Flash 1,084
  58. 58 ERNIE Image Turbo 1,080
  59. 59 ERNIE Image 1,077
  60. 60 Boogu-Image-0.1 1,076
  61. 61 Qwen-Image (2512) 1,072
  62. 62 Cosmos 3 Super 1,065
  63. 63 DALL-E 3 1,049
  64. 64 Vidu Q2 1,043
  65. 65 FLUX.2 [klein] 9B 1,036
  66. 66 Z-Image-Turbo 1,028
  67. 67 Ideogram 3.0 1,022
  68. 68 Nucleus Image 1,017
  69. 69 FLUX.1 [pro] Ultra 1,013
  70. 70 FLUX.2 [klein] 4B 1,005
  71. 71 FLUX.2 Klein 4B Distilled 1,000
  72. 72 Grok 2 Image Gen 1,000
  73. 73 FLUX.1 Krea Dev 990
  74. 74 Imagen 4 Fast Generate 001 963
  75. 75 P-Image 962
  76. 76 Qwen-Image 953
  77. 77 HiDream-O1-Image 937
  78. 78 HiDream O1 Image Dev 903

Source: Official DesignArena Graphic Design leaderboard

#EntryScoreSettingDateSource
1 GPT-Image-2.5 Sunburst 1,436 Official leaderboard
2 GPT-Image-2 1,435 Official leaderboard
3 GPT-Image-2.5 Flare 1,413 Official leaderboard
4 Riverflow 2.5 Pro 1,411 Official leaderboard
5 Reve 2.1 1,372 Official leaderboard
6 Reve 2.0 1,352 Official leaderboard
7 Grok Imagine Image 2.0 1,337 medium Official leaderboard
8 Nano Banana 2.1 1,333 Official leaderboard
9 MAI-Image-2.6 1,330 Official leaderboard
10 Grok Imagine Image 2 Auto 1,314 Official leaderboard
11 MAI-Image-2.6 Flash 1,308 Official leaderboard
12 GPT-Image-1.5 1,295 Official leaderboard
13 Muse Image 1,279 Official leaderboard
14 Seedream 5.0 Pro 1,276 Official leaderboard
15 Qwen-Image-3.0-Pro 1,273 Official leaderboard
16 Gemini 3.1 Flash Image Gen 2K (Nano Banana 2) 1,261 Official leaderboard
17 Ideogram 4.0 1,257 Official leaderboard
18 FLUX 3 Image 1,254 Official leaderboard
19 Gemini 3 Pro Image Gen 2K (Nano Banana Pro) 1,251 Official leaderboard
20 Gemini 3.1 Flash-Lite Image (nano-banana-2-lite) 1,249 Official leaderboard
21 Gemini 3.1 Flash Image Gen (Nano Banana 2) 1,249 Official leaderboard
22 UNI-1.1 Max 1,244 Official leaderboard
23 UNI-1.1 1,242 Official leaderboard
24 Gemini 3 Pro Image Preview 1,226 Official leaderboard
25 MAI-Image-2.5 1,210 Official leaderboard
26 Recraft V4.1 Utility Pro 1,200 Official leaderboard
27 Seedream Lite 5.0 1,190 Official leaderboard
28 FLUX.2 [flex] 1,189 Official leaderboard
29 Imagen 4 Ultra Generate Preview 06-06 1,174 Official leaderboard
30 Recraft V4 Pro 1,173 Official leaderboard
31 FLUX.2 [pro] 1,172 Official leaderboard
32 GPT-Image-1 1,172 Official leaderboard
33 Krea 2 Large 1,171 Official leaderboard
34 Gemini 2.5 Flash Image Gen (Nano Banana) 1,167 Official leaderboard
35 GPT-Image-1 mini 1,165 Official leaderboard
36 Grok Imagine 1,162 Official leaderboard
37 Imagen 4 Ultra (generate-001) 1,159 Official leaderboard
38 ImagineArt 2.0 1,157 Official leaderboard
39 Krea 2 Medium 1,156 Official leaderboard
40 Recraft V4.1 Pro 1,152 Official leaderboard
41 Gemini 2.5 Flash Image (Nano Banana) 1,151 Official leaderboard
42 Imagen 4 Generate Preview 06-06 1,150 Official leaderboard
43 Seedream 4.0 1,148 Official leaderboard
44 p-image-ideogram 1,146 medium, high Official leaderboard
45 Wan 2.7 Pro 1,145 Official leaderboard
46 Wan 2.7 1,143 Official leaderboard
47 Krea 2 Turbo 1,138 Official leaderboard
48 Imagen 4 (generate-001) 1,137 Official leaderboard
49 Recraft V4.1 1,130 Official leaderboard
50 FLUX.1 Kontext [max] 1,128 Official leaderboard
51 GLM-Image 1,109 Official leaderboard
52 Recraft V3 1,096 Official leaderboard
53 Reve Create 1,089 Official leaderboard
54 FLUX.1 Kontext [pro] 1,086 Official leaderboard
55 FLUX.2 [dev] 1,086 Official leaderboard
56 HunyuanImage 3.0 1,085 Official leaderboard
57 Recraft V4.1 Flash 1,084 Official leaderboard
58 ERNIE Image Turbo 1,080 Official leaderboard
59 ERNIE Image 1,077 Official leaderboard
60 Boogu-Image-0.1 1,076 Official leaderboard
61 Qwen-Image (2512) 1,072 Official leaderboard
62 Cosmos 3 Super 1,065 Official leaderboard
63 DALL-E 3 1,049 Official leaderboard
64 Vidu Q2 1,043 Official leaderboard
65 FLUX.2 [klein] 9B 1,036 Official leaderboard
66 Z-Image-Turbo 1,028 Official leaderboard
67 Ideogram 3.0 1,022 Official leaderboard
68 Nucleus Image 1,017 Official leaderboard
69 FLUX.1 [pro] Ultra 1,013 Official leaderboard
70 FLUX.2 [klein] 4B 1,005 Official leaderboard
71 FLUX.2 Klein 4B Distilled 1,000 Official leaderboard
72 Grok 2 Image Gen 1,000 Official leaderboard
73 FLUX.1 Krea Dev 990 Official leaderboard
74 Imagen 4 Fast Generate 001 963 Official leaderboard
75 P-Image 962 Official leaderboard
76 Qwen-Image 953 Official leaderboard
77 HiDream-O1-Image 937 Official leaderboard
78 HiDream O1 Image Dev 903 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

No dated values yet.

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Vision

All 281 in Vision
  • MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. GPT-5.1 85.4%
  • MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
  • CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
  • MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
  • MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy. Qwen 3.8 Max 85.9%
  • ChartQA In use Models answer human-written and generated questions about chart images that need visual and arithmetic reasoning. Scored by relaxed accuracy with a 5 percent numeric tolerance. Command A Vision 90.9%