Benchmark Almanac

SuperCLUE-Writing

Claude Opus 4.7 88.0 leads since April 2026
Not in model cards

Models write original Chinese prose.

SuperCLUE reports scores in a public workbook.

Made by
SuperCLUE
Kind
Creative writing
Level
University
Scored
Graded by people
Answers
Open-ended
Measures
Capability
Language
Chinese
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
A host
Strongest model tested
GPT-5.5 #21 on the capability index
Last model added
May 2026
Scores refreshed
November 20th, 2026
14
Entries on its own leaderboard, 12 models we track

Claude Opus 4.7 leads with 88.0 among models released by May 2026 Claude Opus 4.7 leads with 88.0 among models released by May 2026

Best score per entry from the benchmark's own leaderboard, read on November 20th, 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. This view shows only models we track, each at its best.

The newest model with a score here came out in May 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Claude Opus 4.7 max / API 88.0
  2. 2 ERNIE 5.1 API 87.6
  3. 3 Gemini 3.1 Pro API 86.7
  4. 4 Kimi K2.6 API 86.7
  5. 5 Doubao-Seed-2.0-Pro-260215 API 86.5
  6. 6 Qwen 3.6 Max (Preview) API 86.5
  7. 7 Grok 4.20 reasoning, API 86.2
  8. 8 GLM-5.1 API 85.7
  9. 9 GPT-5.5 API 85.2
  10. 10 DeepSeek-V4-Pro max / API 80.9
  11. 11 MiMo-V2-Pro API 78.9
  12. 12 DeepSeek-V4-Flash max / API 76.6
  13. 13 HY3-Preview(think_high) API 75.3
  14. 14 MiniMax-M2.7 API 56.5

Source: Official SuperCLUE-Writing leaderboard

#EntryScoreSettingDateSource
1 Claude Opus 4.7 88.0 max / API Apr 2026 Official leaderboard
2 ERNIE 5.1 87.6 API May 2026 Official leaderboard
3 Gemini 3.1 Pro 86.7 API Feb 2026 Official leaderboard
4 Kimi K2.6 86.7 API Apr 2026 Official leaderboard
5 Doubao-Seed-2.0-Pro-260215 86.5 API Official leaderboard
6 Qwen 3.6 Max (Preview) 86.5 API Apr 2026 Official leaderboard
7 Grok 4.20 86.2 reasoning, API Feb 2026 Official leaderboard
8 GLM-5.1 85.7 API Apr 2026 Official leaderboard
9 GPT-5.5 85.2 API Apr 2026 Official leaderboard
10 DeepSeek-V4-Pro 80.9 max / API Apr 2026 Official leaderboard
11 MiMo-V2-Pro 78.9 API Mar 2026 Official leaderboard
12 DeepSeek-V4-Flash 76.6 max / API Apr 2026 Official leaderboard
13 HY3-Preview(think_high) 75.3 API Official leaderboard
14 MiniMax-M2.7 56.5 API Mar 2026 Official leaderboard
#ModelScoreSettingDateSource
1 Claude Opus 4.7 88.0 max / API Apr 2026 Official leaderboard
2 ERNIE 5.1 87.6 API May 2026 Official leaderboard
3 Gemini 3.1 Pro 86.7 API Feb 2026 Official leaderboard
4 Kimi K2.6 86.7 API Apr 2026 Official leaderboard
5 Qwen 3.6 Max (Preview) 86.5 API Apr 2026 Official leaderboard
6 Grok 4.20 86.2 reasoning, API Feb 2026 Official leaderboard
7 GLM-5.1 85.7 API Apr 2026 Official leaderboard
8 GPT-5.5 85.2 API Apr 2026 Official leaderboard
9 DeepSeek-V4-Pro 80.9 max / API Apr 2026 Official leaderboard
10 MiMo-V2-Pro 78.9 API Mar 2026 Official leaderboard
11 DeepSeek-V4-Flash 76.6 max / API Apr 2026 Official leaderboard
12 MiniMax-M2.7 56.5 API Mar 2026 Official leaderboard

Rankings by part

We read 26 parts from the benchmark’s own leaderboard in 5 groups: 内容质量, 内容风格, 内容创意, 主流题材创作 and 约束创作. Each ranks models on one part of the test. GPT-5.5 leads 6 of them, and 6 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

内容质量

12 models we track have a score here.

  1. 1 Claude Opus 4.7 90.8
  2. 2 ERNIE 5.1 89.6
  3. 3 Qwen 3.6 Max (Preview) ↑2 86.7
  4. 4 GLM-5.1 ↑3 86.1
  5. 5 Grok 4.20 ↑1 86.1
  6. 6 Gemini 3.1 Pro ↓3 85.0
  7. 7 Kimi K2.6 ↓3 84.1
  8. 8 GPT-5.5 81.3
  9. 9 DeepSeek-V4-Pro 73.8
  10. 10 MiMo-V2-Pro 72.3

3 models have held number one on the official leaderboard since 2023

Lab claims count from the day the card came out, independent runs from the model’s release.

xAI Google DeepMind Anthropic
Own leaderboard Not tested since May 2026

Source: Official SuperCLUE-Writing leaderboard

Scores over time

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

Plot with other benchmarks
On the benchmark’s own leaderboard
50.0 60.0 70.0 80.0 90.0 2026 50.0 60.0 70.0 80.0 90.0 2026

Source: Official SuperCLUE-Writing leaderboard

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 Writing and taste

All 65 in Writing and taste
  • Design Bench (Moonshot) In use Models make graphic designs from direct prompts or few-shot reference images. An automated grader rates the results as Elo scores. Kimi K3.1 Pro 957
  • WritingBench Fading Models complete 1,000 real-world writing requests across 6 domains and 100 subdomains. A judge model scores each reply on five query-specific criteria. GPT-5 83.9
  • Creative Writing (MiniMax in-house) Only in system cards Models write creative pieces for real Hailuo AI user requests. Automatic assessments score the writing. MiniMax-Text-01 81.3%
  • DeepSeek Chinese functional writing evaluation Only in system cards Models answer 3,170 everyday Chinese writing prompts such as reports, emails and speeches. Pairwise judging against a baseline model gives a win rate. DeepSeek-V4-Pro 62.6%
  • Slide generation human evaluation (Z.ai) Only in system cards People compare HTML slide decks from two models on content, layout and visual quality. Metric is overall win rate. GLM-5 67.5%
  • Task-dependent output homogenization evaluation Only in system cards Models answer prompts several times. A judge scores functional diversity where it is wanted and consistency where one answer is right. Muse Spark 0.21

Data audit

7 of 9 checks pass, 2 warnings. 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 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (14 entries).
  • Pass Models scored 12 models we track, plus 2 other entries on its leaderboard.
  • Warning Newest score The newest came 6 months ago, in May 2026, for ERNIE 5.1.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.5 (#21).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on superclueai.com: 12 rows for models we track and 2 other entries at the last read.
  • Pass Last read November 20th, 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.

How we check