Benchmark Almanac

SuperCLUE-CoT

o1 77.7 leads since December 2024
Not in model cards

Models solve Chinese reasoning problems that require several steps.

SuperCLUE reports scores in a public workbook.

Made by
SuperCLUE
Kind
Language reasoning
Level
University
Scored
Checked automatically
Answers
Short answer
Measures
Capability
Language
Chinese
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
A host
Strongest model tested
o1 #118 on the capability index
Last model added
January 2025
Scores refreshed
November 20th, 2026
36
Entries on its own leaderboard, 14 models we track

o1 leads with 77.7 among models released by January 2025 o1 leads with 77.7 among models released by January 2025

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 January 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 o1 API 77.7
  2. 2 o1-preview API 75.0
  3. 3 HaiRuo-General-Reasoning API 68.7
  4. 4 QwQ-32B-Preview API 64.6
  5. 5 DeepSeek-R1 API 63.4
  6. 6 chatgpt-4o-latest API 62.8
  7. 7 Claude-3.5-Sonnet(1022) POE 58.7
  8. 8 Gemini 1.5 Pro Sep 2024, API 56.7
  9. 9 GLM-4-Plus API 53.3
  10. 10 Yi-Lightning API 52.4
  11. 11 GPT-4 Turbo Apr 2024, API 50.8
  12. 12 Qwen-max-0919 API 49.5
  13. 13 Qwen2.5-72B API 47.9
  14. 14 Doubao-pro-32k-240828 API 46.1
  15. 15 o1-mini API 45.5
  16. 16 step-2-16k API 43.8
  17. 17 ERNIE-4.0-Turbo-8K-Preview API 40.8
  18. 18 Llama3.1-405B-Instruct API 40.7
  19. 19 Minimax-abab6.5s-chat API 40.6
  20. 20 hunyuan-turbo-latest API 39.9
  21. 21 kimi.ai(网页) 网页 38.0
  22. 22 Llama3.1-70B-Instruct API 37.7
  23. 23 gemma-2-9b-it 模型 36.2
  24. 24 Qwen2.5-7b-instruct API 35.9
  25. 25 Spark-4-Ultra API 34.2
  26. 26 Baichuan4 API 33.9
  27. 27 Yi-1.5-9b-chat-16k 模型 31.3
  28. 28 GLM-4-9B-Chat 模型 31.1
  29. 29 Yi-1.5-6B-Chat 模型 27.5
  30. 30 Llama3.1-8B-Instruct API 24.0
  31. 31 Gemma 2 2B IT 模型 22.1
  32. 32 Qwen2.5-1.5b-instruct API 21.2
  33. 33 phi-3-mini 3.8B 模型 20.6
  34. 34 Mistral 7B v0.3 模型 16.6
  35. 35 Qwen2.5-0.5b-instruct API 11.9
  36. 36 RWKV-6-World-7b 模型 11.5

Source: Official SuperCLUE-CoT leaderboard

#EntryScoreSettingReleasedRun onSource
1 o1 77.7 API Dec 2024 Apr 2025 Official leaderboard
2 o1-preview 75.0 API Sep 2024 Apr 2025 Official leaderboard
3 HaiRuo-General-Reasoning 68.7 API Apr 2025 Official leaderboard
4 QwQ-32B-Preview 64.6 API Nov 2024 Apr 2025 Official leaderboard
5 DeepSeek-R1 63.4 API Jan 2025 Apr 2025 Official leaderboard
6 chatgpt-4o-latest 62.8 API Apr 2025 Official leaderboard
7 Claude-3.5-Sonnet(1022) 58.7 POE Apr 2025 Official leaderboard
8 Gemini 1.5 Pro Sep 2024 56.7 API Sep 2024 Apr 2025 Official leaderboard
9 GLM-4-Plus 53.3 API Apr 2025 Official leaderboard
10 Yi-Lightning 52.4 API Dec 2024 Apr 2025 Official leaderboard
11 GPT-4 Turbo Apr 2024 50.8 API Apr 2024 Apr 2025 Official leaderboard
12 Qwen-max-0919 49.5 API Apr 2025 Official leaderboard
13 Qwen2.5-72B 47.9 API Sep 2024 Apr 2025 Official leaderboard
14 Doubao-pro-32k-240828 46.1 API Apr 2025 Official leaderboard
15 o1-mini 45.5 API Sep 2024 Apr 2025 Official leaderboard
16 step-2-16k 43.8 API Apr 2025 Official leaderboard
17 ERNIE-4.0-Turbo-8K-Preview 40.8 API Apr 2025 Official leaderboard
18 Llama3.1-405B-Instruct 40.7 API Apr 2025 Official leaderboard
19 Minimax-abab6.5s-chat 40.6 API Apr 2025 Official leaderboard
20 hunyuan-turbo-latest 39.9 API Apr 2025 Official leaderboard
21 kimi.ai(网页) 38.0 网页 Apr 2025 Official leaderboard
22 Llama3.1-70B-Instruct 37.7 API Apr 2025 Official leaderboard
23 gemma-2-9b-it 36.2 模型 Apr 2025 Official leaderboard
24 Qwen2.5-7b-instruct 35.9 API Apr 2025 Official leaderboard
25 Spark-4-Ultra 34.2 API Apr 2025 Official leaderboard
26 Baichuan4 33.9 API Apr 2025 Official leaderboard
27 Yi-1.5-9b-chat-16k 31.3 模型 Apr 2025 Official leaderboard
28 GLM-4-9B-Chat 31.1 模型 Jun 2024 Apr 2025 Official leaderboard
29 Yi-1.5-6B-Chat 27.5 模型 May 2024 Apr 2025 Official leaderboard
30 Llama3.1-8B-Instruct 24.0 API Apr 2025 Official leaderboard
31 Gemma 2 2B IT 22.1 模型 Jul 2024 Apr 2025 Official leaderboard
32 Qwen2.5-1.5b-instruct 21.2 API Apr 2025 Official leaderboard
33 phi-3-mini 3.8B 20.6 模型 Apr 2024 Apr 2025 Official leaderboard
34 Mistral 7B v0.3 16.6 模型 Sep 2023 Apr 2025 Official leaderboard
35 Qwen2.5-0.5b-instruct 11.9 API Apr 2025 Official leaderboard
36 RWKV-6-World-7b 11.5 模型 Apr 2025 Official leaderboard
#ModelScoreSettingReleasedRun onSource
1 o1 77.7 API Dec 2024 Apr 2025 Official leaderboard
2 o1-preview 75.0 API Sep 2024 Apr 2025 Official leaderboard
3 QwQ-32B-Preview 64.6 API Nov 2024 Apr 2025 Official leaderboard
4 DeepSeek-R1 63.4 API Jan 2025 Apr 2025 Official leaderboard
5 Gemini 1.5 Pro Sep 2024 56.7 API Sep 2024 Apr 2025 Official leaderboard
6 Yi-Lightning 52.4 API Dec 2024 Apr 2025 Official leaderboard
7 GPT-4 Turbo Apr 2024 50.8 API Apr 2024 Apr 2025 Official leaderboard
8 Qwen2.5-72B 47.9 API Sep 2024 Apr 2025 Official leaderboard
9 o1-mini 45.5 API Sep 2024 Apr 2025 Official leaderboard
10 GLM-4-9B-Chat 31.1 模型 Jun 2024 Apr 2025 Official leaderboard
11 Yi-1.5-6B-Chat 27.5 模型 May 2024 Apr 2025 Official leaderboard
12 Gemma 2 2B IT 22.1 模型 Jul 2024 Apr 2025 Official leaderboard
13 phi-3-mini 3.8B 20.6 模型 Apr 2024 Apr 2025 Official leaderboard
14 Mistral 7B v0.3 16.6 模型 Sep 2023 Apr 2025 Official leaderboard

Rankings by part

We read 7 parts from the benchmark’s own leaderboard in 2 groups: 回答质量 and 思维链与反思. Each ranks models on one part of the test. o1 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.

回答质量

14 models we track have a score here.

  1. 1 o1 90.5
  2. 2 o1-preview 88.4
  3. 3 DeepSeek-R1 ↑1 80.8
  4. 4 QwQ-32B-Preview ↓1 68.3
  5. 5 Gemini 1.5 Pro Sep 2024 55.0
  6. 6 GPT-4 Turbo Apr 2024 ↑1 51.3
  7. 7 Yi-Lightning ↓1 49.5
  8. 8 o1-mini ↑1 43.6
  9. 9 Qwen2.5-72B ↓1 43.4
  10. 10 GLM-4-9B-Chat 25.4

4 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.

Mistral AI OpenAI
Own leaderboard Not tested since Jan 2025
o1 77.7

Source: Official SuperCLUE-CoT 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
0.00 20.0 40.0 60.0 80.0 2024 2025 0.00 20.0 40.0 60.0 80.0 2024 2025

Source: Official SuperCLUE-CoT 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 Knowledge

All 332 in Knowledge
  • MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects. Intern-S2-397B 89.8%
  • MMLU In use 57-subject multiple-choice exam questions from elementary to professional level. Scored by accuracy. Seed1.8 92.3%
  • SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy. Qwen 3.6 Max (Preview) 73.9%
  • WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1. Tencent Hy4 preview 83.9%
  • BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly. Gemma 4 31B IT 74.4%
  • HLE-Diamond In use Models answer 1,000 cleaned expert questions from Humanity's Last Exam, 500 on reasoning and 500 on knowledge, closed book. Graded on exact answers. Claude Fable 5.2 77.9%

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 (36 entries).
  • Pass Models scored 14 models we track, plus 22 other entries on its leaderboard.
  • Warning Newest score The newest came 22 months ago, in January 2025, for DeepSeek-R1.
  • Note Frontier models None of the top 10 has a score. The strongest tested is o1 (#118).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on superclueai.com: 14 rows for models we track and 22 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