Benchmark Almanac

StudyArena Image

Riverflow 2.5 Pro 1,349 tops its leaderboard Grok Imagine Image 2.0 1,326 leads the models we track
Not in model cards

Models produce image responses to user prompts. Anonymous human comparisons determine the published rating in this separate vote pool.

Only entries with observed ballots count as results. Topic filters do not form separate benchmark records.

Made by
StudyArena
Kind
Image generation and editing
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Scores refreshed
November 12th, 2026
40
Entries on its own leaderboard, 23 models we track

Riverflow 2.5 Pro tops the leaderboard with 1,349

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. The best model we track is Grok Imagine Image 2.0, at 1,326.

40 entries. Scroll sideways for the rest. Bars start at 1,000.

  1. 1,349 Riverflow 2.5 Pro
  2. 1,326 Grok Imagine Image 2.0
  3. 1,320 GPT-Image-2.5 Flare
  4. 1,299 FLUX.3 Image
  5. 1,289 Seedream 5.0 Pro
  6. 1,285 MAI Image 2.6 Flash
  7. 1,271 Seedream 5.0 Flash
  8. 1,270 Gemini 3 Pro Image
  9. 1,253 MAI-Image-2.6
  10. 1,239 GPT-Image-2.5 Sunburst
  11. 1,223 Gemini 3.1 Flash Image
  12. 1,223 Nano Banana 2.1
  13. 1,218 Grok Imagine Quality
  14. 1,216 GPT-Image-2
  15. 1,215 Z-Image-Turbo
  16. 1,215 Qwen Image 3
  17. 1,215 Recraft V4.1
  18. 1,213 Gemini 2.5 Flash Image
  19. 1,212 FLUX.2 [pro]
  20. 1,210 Riverflow 2.5 Fast
  21. 1,205 GPT-Image-1 mini
  22. 1,205 Ideogram 4 Fast
  23. 1,200 MAI Image 2.5 Pro
  24. 1,196 GPT-Image-1
  25. 1,179 Recraft V4.1 Pro
  26. 1,177 FLUX.2 [max]
  27. 1,170 Krea 2 Medium Turbo
  28. 1,159 Krea 2 Large
  29. 1,154 Seedream 5.0 Lite
  30. 1,145 Ideogram 4.0
  31. 1,142 FLUX.2 [flex]
  32. 1,138 Gemini 3.1 Flash Lite Image
  33. 1,131 Muse Image
  34. 1,129 Seedream 4.5
  35. 1,127 MAI-Image-2.5
  36. 1,122 Recraft V4.1 Utility
  37. 1,119 Krea 2 Medium
  38. 1,107 Qwen-Image-3.0-Pro
  39. 1,088 FLUX.2 [klein] 4B
  40. 1,047 Recraft V4.1 Utility Pro

Bars start at 1,000.

  1. 1 Riverflow 2.5 Pro 1,349
  2. 2 Grok Imagine Image 2.0 1,326
  3. 3 GPT-Image-2.5 Flare 1,320
  4. 4 FLUX.3 Image 1,299
  5. 5 Seedream 5.0 Pro 1,289
  6. 6 MAI Image 2.6 Flash 1,285
  7. 7 Seedream 5.0 Flash 1,271
  8. 8 Gemini 3 Pro Image 1,270
  9. 9 MAI-Image-2.6 1,253
  10. 10 GPT-Image-2.5 Sunburst 1,239
  11. 11 Gemini 3.1 Flash Image 1,223
  12. 12 Nano Banana 2.1 1,223
  13. 13 Grok Imagine Quality 1,218
  14. 14 GPT-Image-2 1,216
  15. 15 Z-Image-Turbo 1,215
  16. 16 Qwen Image 3 1,215
  17. 17 Recraft V4.1 1,215
  18. 18 Gemini 2.5 Flash Image 1,213
  19. 19 FLUX.2 [pro] 1,212
  20. 20 Riverflow 2.5 Fast 1,210
  21. 21 GPT-Image-1 mini 1,205
  22. 22 Ideogram 4 Fast 1,205
  23. 23 MAI Image 2.5 Pro 1,200
  24. 24 GPT-Image-1 1,196
  25. 25 Recraft V4.1 Pro 1,179
  26. 26 FLUX.2 [max] 1,177
  27. 27 Krea 2 Medium Turbo 1,170
  28. 28 Krea 2 Large 1,159
  29. 29 Seedream 5.0 Lite 1,154
  30. 30 Ideogram 4.0 1,145
  31. 31 FLUX.2 [flex] 1,142
  32. 32 Gemini 3.1 Flash Lite Image 1,138
  33. 33 Muse Image 1,131
  34. 34 Seedream 4.5 1,129
  35. 35 MAI-Image-2.5 1,127
  36. 36 Recraft V4.1 Utility 1,122
  37. 37 Krea 2 Medium 1,119
  38. 38 Qwen-Image-3.0-Pro 1,107
  39. 39 FLUX.2 [klein] 4B 1,088
  40. 40 Recraft V4.1 Utility Pro 1,047

Source: Official StudyArena Image leaderboard

#EntryScoreSettingDateSource
1 Riverflow 2.5 Pro 1,349 Official leaderboard
2 Grok Imagine Image 2.0 1,326 Official leaderboard
3 GPT-Image-2.5 Flare 1,320 Official leaderboard
4 FLUX.3 Image 1,299 Official leaderboard
5 Seedream 5.0 Pro 1,289 Official leaderboard
6 MAI Image 2.6 Flash 1,285 Official leaderboard
7 Seedream 5.0 Flash 1,271 Official leaderboard
8 Gemini 3 Pro Image 1,270 Official leaderboard
9 MAI-Image-2.6 1,253 Official leaderboard
10 GPT-Image-2.5 Sunburst 1,239 Official leaderboard
11 Gemini 3.1 Flash Image 1,223 Official leaderboard
12 Nano Banana 2.1 1,223 Official leaderboard
13 Grok Imagine Quality 1,218 Official leaderboard
14 GPT-Image-2 1,216 Official leaderboard
15 Z-Image-Turbo 1,215 Official leaderboard
16 Qwen Image 3 1,215 Official leaderboard
17 Recraft V4.1 1,215 Official leaderboard
18 Gemini 2.5 Flash Image 1,213 Official leaderboard
19 FLUX.2 [pro] 1,212 Official leaderboard
20 Riverflow 2.5 Fast 1,210 Official leaderboard
21 GPT-Image-1 mini 1,205 Official leaderboard
22 Ideogram 4 Fast 1,205 Official leaderboard
23 MAI Image 2.5 Pro 1,200 Official leaderboard
24 GPT-Image-1 1,196 Official leaderboard
25 Recraft V4.1 Pro 1,179 Official leaderboard
26 FLUX.2 [max] 1,177 Official leaderboard
27 Krea 2 Medium Turbo 1,170 Official leaderboard
28 Krea 2 Large 1,159 Official leaderboard
29 Seedream 5.0 Lite 1,154 Official leaderboard
30 Ideogram 4.0 1,145 Official leaderboard
31 FLUX.2 [flex] 1,142 Official leaderboard
32 Gemini 3.1 Flash Lite Image 1,138 Official leaderboard
33 Muse Image 1,131 Official leaderboard
34 Seedream 4.5 1,129 Official leaderboard
35 MAI-Image-2.5 1,127 Official leaderboard
36 Recraft V4.1 Utility 1,122 Official leaderboard
37 Krea 2 Medium 1,119 Official leaderboard
38 Qwen-Image-3.0-Pro 1,107 Official leaderboard
39 FLUX.2 [klein] 4B 1,088 Official leaderboard
40 Recraft V4.1 Utility Pro 1,047 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%