Benchmark Almanac

SuperCLUE-Math6o

DeepSeek-R1 82.2 leads since January 2025
Not in model cards

Models solve Chinese elementary school math olympiad problems.

SuperCLUE reports scores in a public workbook.

Made by
SuperCLUE
Kind
Competition problems with answers
Level
Competition
Scored
Checked automatically
Answers
Short answer
Measures
Capability
Subject
Mathematics
Language
Chinese
In and out
Text → Text
Interaction
Single turn
Items from
Exams and contests
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
37
Entries on its own leaderboard, 14 models we track

DeepSeek-R1 leads with 82.2 among models released by January 2025 DeepSeek-R1 leads with 82.2 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 DeepSeek-R1 API 82.2
  2. 2 o1 网页 76.0
  3. 3 o1-preview API 73.7
  4. 4 DeepSeek-R1-Lite-Preview 网页 69.9
  5. 5 QwQ-32B-Preview API 66.8
  6. 6 k0-math 网页 57.4
  7. 7 chatgpt-4o-latest API 57.3
  8. 8 Gemini 1.5 Pro Sep 2024, API 55.5
  9. 9 InternThinker 网页 54.8
  10. 10 Doubao-pro-32k-240828 API 54.8
  11. 11 Qwen2.5-72B API 50.4
  12. 12 Qwen-max-0919 API 49.8
  13. 13 Baichuan4-Turbo API 49.3
  14. 14 kimi.ai(网页) 网页 48.7
  15. 15 Claude-3.5-Sonnet(1022) POE 47.4
  16. 16 DeepSeek-V2.5 Sep 2024, API 47.3
  17. 17 step-2-16k API 46.4
  18. 18 GLM-4-Plus API 46.0
  19. 19 o1-mini API 44.0
  20. 20 GPT-4 Turbo Apr 2024, API 43.4
  21. 21 Qwen2.5-7b-instruct API 39.9
  22. 22 Spark-4-Ultra API 38.4
  23. 23 Llama3.1-405B-Instruct API 37.2
  24. 24 MiniMax-abab6.5s-chat API 34.5
  25. 25 Baichuan4 API 33.1
  26. 26 GLM-4-9B-Chat 模型 31.6
  27. 27 Llama3.1-70B-Instruct API 30.9
  28. 28 gemma-2-9b-it 模型 29.3
  29. 29 Yi-1.5-6B-Chat 模型 25.1
  30. 30 Yi-1.5-9b-chat-16k 模型 24.8
  31. 31 Qwen2.5-1.5b-instruct API 23.8
  32. 32 Llama3.1-8B-Instruct API 23.8
  33. 33 Gemma 2 2B IT 模型 21.6
  34. 34 phi-3-mini 3.8B 模型 19.8
  35. 35 Qwen2.5-0.5b-instruct API 18.8
  36. 36 Mistral 7B v0.3 模型 17.6
  37. 37 RWKV-6-World-7b 模型 15.3

Source: Official SuperCLUE-Math6o leaderboard

#EntryScoreSettingReleasedRun onSource
1 DeepSeek-R1 82.2 API Jan 2025 Jan 2025 Official leaderboard
2 o1 76.0 网页 Dec 2024 Jan 2025 Official leaderboard
3 o1-preview 73.7 API Sep 2024 Jan 2025 Official leaderboard
4 DeepSeek-R1-Lite-Preview 69.9 网页 Jan 2025 Official leaderboard
5 QwQ-32B-Preview 66.8 API Nov 2024 Jan 2025 Official leaderboard
6 k0-math 57.4 网页 Jan 2025 Official leaderboard
7 chatgpt-4o-latest 57.3 API Jan 2025 Official leaderboard
8 Gemini 1.5 Pro Sep 2024 55.5 API Sep 2024 Jan 2025 Official leaderboard
9 InternThinker 54.8 网页 Jan 2025 Official leaderboard
10 Doubao-pro-32k-240828 54.8 API Jan 2025 Official leaderboard
11 Qwen2.5-72B 50.4 API Sep 2024 Jan 2025 Official leaderboard
12 Qwen-max-0919 49.8 API Jan 2025 Official leaderboard
13 Baichuan4-Turbo 49.3 API Jan 2025 Official leaderboard
14 kimi.ai(网页) 48.7 网页 Jan 2025 Official leaderboard
15 Claude-3.5-Sonnet(1022) 47.4 POE Jan 2025 Official leaderboard
16 DeepSeek-V2.5 Sep 2024 47.3 API Sep 2024 Jan 2025 Official leaderboard
17 step-2-16k 46.4 API Jan 2025 Official leaderboard
18 GLM-4-Plus 46.0 API Jan 2025 Official leaderboard
19 o1-mini 44.0 API Sep 2024 Jan 2025 Official leaderboard
20 GPT-4 Turbo Apr 2024 43.4 API Apr 2024 Jan 2025 Official leaderboard
21 Qwen2.5-7b-instruct 39.9 API Jan 2025 Official leaderboard
22 Spark-4-Ultra 38.4 API Jan 2025 Official leaderboard
23 Llama3.1-405B-Instruct 37.2 API Jan 2025 Official leaderboard
24 MiniMax-abab6.5s-chat 34.5 API Jan 2025 Official leaderboard
25 Baichuan4 33.1 API Jan 2025 Official leaderboard
26 GLM-4-9B-Chat 31.6 模型 Jun 2024 Jan 2025 Official leaderboard
27 Llama3.1-70B-Instruct 30.9 API Jan 2025 Official leaderboard
28 gemma-2-9b-it 29.3 模型 Jan 2025 Official leaderboard
29 Yi-1.5-6B-Chat 25.1 模型 May 2024 Jan 2025 Official leaderboard
30 Yi-1.5-9b-chat-16k 24.8 模型 Jan 2025 Official leaderboard
31 Qwen2.5-1.5b-instruct 23.8 API Jan 2025 Official leaderboard
32 Llama3.1-8B-Instruct 23.8 API Jan 2025 Official leaderboard
33 Gemma 2 2B IT 21.6 模型 Jul 2024 Jan 2025 Official leaderboard
34 phi-3-mini 3.8B 19.8 模型 Apr 2024 Jan 2025 Official leaderboard
35 Qwen2.5-0.5b-instruct 18.8 API Jan 2025 Official leaderboard
36 Mistral 7B v0.3 17.6 模型 Sep 2023 Jan 2025 Official leaderboard
37 RWKV-6-World-7b 15.3 模型 Jan 2025 Official leaderboard
#ModelScoreSettingReleasedRun onSource
1 DeepSeek-R1 82.2 API Jan 2025 Jan 2025 Official leaderboard
2 o1 76.0 网页 Dec 2024 Jan 2025 Official leaderboard
3 o1-preview 73.7 API Sep 2024 Jan 2025 Official leaderboard
4 QwQ-32B-Preview 66.8 API Nov 2024 Jan 2025 Official leaderboard
5 Gemini 1.5 Pro Sep 2024 55.5 API Sep 2024 Jan 2025 Official leaderboard
6 Qwen2.5-72B 50.4 API Sep 2024 Jan 2025 Official leaderboard
7 DeepSeek-V2.5 Sep 2024 47.3 API Sep 2024 Jan 2025 Official leaderboard
8 o1-mini 44.0 API Sep 2024 Jan 2025 Official leaderboard
9 GPT-4 Turbo Apr 2024 43.4 API Apr 2024 Jan 2025 Official leaderboard
10 GLM-4-9B-Chat 31.6 模型 Jun 2024 Jan 2025 Official leaderboard
11 Yi-1.5-6B-Chat 25.1 模型 May 2024 Jan 2025 Official leaderboard
12 Gemma 2 2B IT 21.6 模型 Jul 2024 Jan 2025 Official leaderboard
13 phi-3-mini 3.8B 19.8 模型 Apr 2024 Jan 2025 Official leaderboard
14 Mistral 7B v0.3 17.6 模型 Sep 2023 Jan 2025 Official leaderboard

Rankings by part

We read 7 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as 行程, 数论 and 几何. DeepSeek-R1 leads 5 of them, and o1 leads the other 2. Arrows show how far a model moved from its overall rank among the models we track.

行程

14 models we track have a score here.

  1. 1 DeepSeek-R1 82.0
  2. 2 o1-preview ↑1 78.1
  3. 3 o1 ↓1 58.7
  4. 4 QwQ-32B-Preview 56.8
  5. 5 Gemini 1.5 Pro Sep 2024 52.1
  6. 6 Qwen2.5-72B 49.3
  7. 7 DeepSeek-V2.5 Sep 2024 40.4
  8. 8 GPT-4 Turbo Apr 2024 ↑1 38.8
  9. 9 o1-mini ↓1 29.4
  10. 10 Yi-1.5-6B-Chat ↑1 26.7

6 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 DeepSeek
Own leaderboard Not tested since Jan 2025

Source: Official SuperCLUE-Math6o 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 80.0 100.0 2024 2025 0.00 20.0 40.0 60.0 80.0 100.0 2024 2025

Source: Official SuperCLUE-Math6o 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 Math

All 140 in Math
  • AIME 2026 In use Models solve the 30 integer-answer problems from the 2026 AIME I and II math competitions. Scored by exact answer match. GPT-5.5 100%
  • IMO-AnswerBench In use 400 olympiad problems with short verifiable answers, robustified against memorization. Scored by answer match. GLM-5.2 91%
  • HMMT 2026 In use Models solve the 33 final-answer problems from the February 2026 Harvard-MIT Mathematics Tournament. Scored by exact answer match. GPT-5.5 98.5%
  • PolyMath In use Models solve 9,000 parallel math problems in 18 languages across four difficulty levels, from school math to frontier math. Final answers are checked. Qwen3.7-Max 86.5%
  • MathArena (Apex, IMO, USAMO) In use Evaluates models on newly released math competitions after release. Final-answer or human/LLM-judged proof scores. Claude Opus 4.8 81.3%
  • MathArena Apex Shortlist In use Models solve 47 hard final-answer problems that MathArena selected from 2025 competitions, where the best models scored about 50% at selection. Scored by accuracy over repeated attempts. GPT-5.5 98.4%

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 (37 entries).
  • Pass Models scored 14 models we track, plus 23 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 23 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