SuperCLUE-Image
Models generate images from Chinese prompts. The benchmark scores text rendering, visual quality, prompt adherence and scene composition.
SuperCLUE publishes scores for 25 models in its 2026 results workbook.
- Made by
- SuperCLUE
- Released
- June 2024
- Kind
- Image generation and editing
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Open-ended
- Measures
- Capability
- Language
- Chinese
- Interaction
- Single turn
- Refresh
- New versions replace it
- Run by
- A host
- In views
- Image, video and 3D generation
- Scores refreshed
- November 21st, 2026
ArtArk tops the leaderboard with 95.1 GPT-Image-2.5 Sunburst leads with 94.3
Best score per entry from the benchmark's own leaderboard, read on November 21st, 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 GPT-Image-2.5 Sunburst, at 94.3. This view shows only models we track, each at its best.
- 95.1 ArtArk
- 94.3 GPT-Image-2.5 Sunburst
- 93.8 SenseNova U1 Pro
- 93.5 GPT-Image-2.5 Flare
- 93.2 GPT-Image-2
- 92.9 Nano Banana 2.1
- 91.9 Doubao-Seedream-5.0-pro
- 91.6
Qwen-Image-3.0-Pro
- 91.2
MAI-Image-2.6
- 90.3 Grok Imagine Image 2.0
- 90.2 SenseNova U1.5 Lite
- 90.1 Gemini-3.1-Flash-Image (Nano Banana 2)
- 89.7 JT4.1-Image
- 89.4 Doubao-Seedream-5.0-lite
- 86.7
Qwen-Image-2.0-Pro
- 84.9 Nano Banana Pro
- 84.7 Doubao-Seedream-4.5
- 77.8 Hy-Image-3.0
- 76.3 Recraft V4.1 Utility Pro
- 75.2 Wan2.6-t2i
- 66.0 Kling-v3
- 63.1 Kling-v2-1
- 61.9 FLUX.2 [max]
- 58.9 FLUX.2 [pro]
- 46.3 CogView-4
- 1 ArtArk 95.1
- 2 GPT-Image-2.5 Sunburst 94.3
- 3 SenseNova U1 Pro 93.8
- 4 GPT-Image-2.5 Flare 93.5
- 5 GPT-Image-2 93.2
- 6 Nano Banana 2.1 92.9
- 7 Doubao-Seedream-5.0-pro 91.9
- 8
Qwen-Image-3.0-Pro 91.6
- 9
MAI-Image-2.6 91.2
- 10 Grok Imagine Image 2.0 90.3
- 11 SenseNova U1.5 Lite 90.2
- 12 Gemini-3.1-Flash-Image (Nano Banana 2) 90.1
- 13 JT4.1-Image 89.7
- 14 Doubao-Seedream-5.0-lite 89.4
- 15
Qwen-Image-2.0-Pro 86.7
- 16 Nano Banana Pro 84.9
- 17 Doubao-Seedream-4.5 84.7
- 18 Hy-Image-3.0 77.8
- 19 Recraft V4.1 Utility Pro 76.3
- 20 Wan2.6-t2i 75.2
- 21 Kling-v3 66.0
- 22 Kling-v2-1 63.1
- 23 FLUX.2 [max] 61.9
- 24 FLUX.2 [pro] 58.9
- 25 CogView-4 46.3
Source: Official SuperCLUE-Image leaderboard
- 1 GPT-Image-2.5 Sunburst 94.3
- 2 GPT-Image-2.5 Flare 93.5
- 3 GPT-Image-2 93.2
- 4
Qwen-Image-3.0-Pro 91.6
- 5
MAI-Image-2.6 91.2
- 6 Grok Imagine Image 2.0 90.3
- 7
Qwen-Image-2.0-Pro 86.7
- 8 Recraft V4.1 Utility Pro 76.3
- 9 FLUX.2 [max] 61.9
- 10 FLUX.2 [pro] 58.9
Source: Official SuperCLUE-Image leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | ArtArk | 95.1 | Official leaderboard | ||
| 2 | GPT-Image-2.5 Sunburst | 94.3 | Official leaderboard | ||
| 3 | SenseNova U1 Pro | 93.8 | Official leaderboard | ||
| 4 | GPT-Image-2.5 Flare | 93.5 | Official leaderboard | ||
| 5 | GPT-Image-2 | 93.2 | Official leaderboard | ||
| 6 | Nano Banana 2.1 | 92.9 | Official leaderboard | ||
| 7 | Doubao-Seedream-5.0-pro | 91.9 | Official leaderboard | ||
| 8 | | 91.6 | Official leaderboard | ||
| 9 | | 91.2 | Official leaderboard | ||
| 10 | Grok Imagine Image 2.0 | 90.3 | Official leaderboard | ||
| 11 | SenseNova U1.5 Lite | 90.2 | Official leaderboard | ||
| 12 | Gemini-3.1-Flash-Image (Nano Banana 2) | 90.1 | Official leaderboard | ||
| 13 | JT4.1-Image | 89.7 | Official leaderboard | ||
| 14 | Doubao-Seedream-5.0-lite | 89.4 | Official leaderboard | ||
| 15 | | 86.7 | Official leaderboard | ||
| 16 | Nano Banana Pro | 84.9 | Official leaderboard | ||
| 17 | Doubao-Seedream-4.5 | 84.7 | Official leaderboard | ||
| 18 | Hy-Image-3.0 | 77.8 | Official leaderboard | ||
| 19 | Recraft V4.1 Utility Pro | 76.3 | Official leaderboard | ||
| 20 | Wan2.6-t2i | 75.2 | Official leaderboard | ||
| 21 | Kling-v3 | 66.0 | Official leaderboard | ||
| 22 | Kling-v2-1 | 63.1 | Official leaderboard | ||
| 23 | FLUX.2 [max] | 61.9 | Official leaderboard | ||
| 24 | FLUX.2 [pro] | 58.9 | Official leaderboard | ||
| 25 | CogView-4 | 46.3 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-Image-2.5 Sunburst | 94.3 | Official leaderboard | ||
| 2 | GPT-Image-2.5 Flare | 93.5 | Official leaderboard | ||
| 3 | GPT-Image-2 | 93.2 | Official leaderboard | ||
| 4 | | 91.6 | Official leaderboard | ||
| 5 | | 91.2 | Official leaderboard | ||
| 6 | Grok Imagine Image 2.0 | 90.3 | Official leaderboard | ||
| 7 | | 86.7 | Official leaderboard | ||
| 8 | Recraft V4.1 Utility Pro | 76.3 | Official leaderboard | ||
| 9 | FLUX.2 [max] | 61.9 | Official leaderboard | ||
| 10 | FLUX.2 [pro] | 58.9 | Official leaderboard |
Rankings by part
We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as 图像质量, 图文一致性 and 汉字生成. GPT-Image-2.5 Sunburst leads 3 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
图像质量
10 models we track have a score here.
- 1
MAI-Image-2.6 ↑4 91.4
- 2 GPT-Image-2.5 Flare 90.8
- 3 GPT-Image-2.5 Sunburst ↓2 90.0
- 4 GPT-Image-2 ↓1 89.8
- 5 Grok Imagine Image 2.0 ↑1 87.4
- 6
Qwen-Image-2.0-Pro ↑1 87.0
- 7
Qwen-Image-3.0-Pro ↓3 86.6
- 8 FLUX.2 [max] ↑1 84.0
- 9 Recraft V4.1 Utility Pro ↓1 80.0
- 10 FLUX.2 [pro] 79.4
图文一致性
10 models we track have a score here.
- 1 GPT-Image-2.5 Flare ↑1 93.9
- 2
MAI-Image-2.6 ↑3 92.4
- 3 GPT-Image-2.5 Sunburst ↓2 90.9
- 4 GPT-Image-2 ↓1 87.8
- 5 Grok Imagine Image 2.0 ↑1 81.8
- 6
Qwen-Image-2.0-Pro ↑1 81.8
- 7
Qwen-Image-3.0-Pro ↓3 81.8
- 8 Recraft V4.1 Utility Pro 81.8
- 9 FLUX.2 [max] 65.2
- 10 FLUX.2 [pro] 54.5
汉字生成
10 models we track have a score here.
- 1 GPT-Image-2.5 Sunburst 97.3
- 2
Qwen-Image-3.0-Pro ↑2 97.0
- 3 GPT-Image-2 94.9
- 4 Grok Imagine Image 2.0 ↑2 93.3
- 5 GPT-Image-2.5 Flare ↓3 92.0
- 6
MAI-Image-2.6 ↓1 85.7
- 7
Qwen-Image-2.0-Pro 82.8
- 8 Recraft V4.1 Utility Pro 52.1
- 9 FLUX.2 [max] 45.2
- 10 FLUX.2 [pro] 42.8
现实复现
10 models we track have a score here.
- 1 GPT-Image-2.5 Sunburst 96.5
- 2 GPT-Image-2.5 Flare 96.1
- 3 GPT-Image-2 96.0
- 4
Qwen-Image-3.0-Pro 95.1
- 5 Grok Imagine Image 2.0 ↑1 94.7
- 6
Qwen-Image-2.0-Pro ↑1 92.9
- 7 Recraft V4.1 Utility Pro ↑1 90.5
- 8
MAI-Image-2.6 ↓3 89.7
- 9 FLUX.2 [max] 65.5
- 10 FLUX.2 [pro] 65.0
创作与推理
10 models we track have a score here.
- 1 GPT-Image-2 ↑2 96.8
- 2 GPT-Image-2.5 Sunburst ↓1 96.6
- 3 GPT-Image-2.5 Flare ↓1 96.0
- 4
Qwen-Image-3.0-Pro 95.8
- 5
MAI-Image-2.6 95.7
- 6 Grok Imagine Image 2.0 95.6
- 7
Qwen-Image-2.0-Pro 91.3
- 8 Recraft V4.1 Utility Pro 86.8
- 9 FLUX.2 [pro] ↑1 84.6
- 10 FLUX.2 [max] ↓1 83.0
复杂信息布局
10 models we track have a score here.
- 1 GPT-Image-2.5 Sunburst 94.3
- 2 GPT-Image-2 ↑1 94.0
- 3
Qwen-Image-3.0-Pro ↑1 93.3
- 4
MAI-Image-2.6 ↑1 92.0
- 5 GPT-Image-2.5 Flare ↓3 91.9
- 6 Grok Imagine Image 2.0 89.1
- 7
Qwen-Image-2.0-Pro 84.5
- 8 Recraft V4.1 Utility Pro 66.3
- 9 FLUX.2 [max] 28.6
- 10 FLUX.2 [pro] 27.1
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 540 launch posts from 57 labs, and none of them list this benchmark.
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy.
Qwen 3.6 Plus 86%
- 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%
- MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy.
Seed2.1 Pro 90.7%
- 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%
Data audit
7 of 8 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (25 entries).
- Pass Models scored 10 models we track, plus 15 other entries on its leaderboard.
- Note Newest score No scores yet.
- Note Frontier models None of its models are on the capabilities index.
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on superclueai.com: 10 rows for models we track and 15 other entries at the last read.
- Pass Last read November 21st, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.