Benchmark Almanac

SuperCLUE Arena Web Coding

Qwen3.8 Max (0902) 1,462 leads since September 2026
Not in model cards

Models implement web interfaces. The arena ranks systems from pairwise comparisons.

The arena reports median ratings and uncertainty intervals.

Made by
SuperCLUE
Kind
Web and app front ends
Level
University
Scored
Ranked by matches
Answers
Code
Tasks
All public
Measures
Preference
Language
Chinese
In and out
Text → Code
Interaction
Single turn
Items from
Live play or votes
Refresh
Live
Run by
A host
In views
Kept fresh
Strongest model tested
Kimi K3.1 Pro #17 on the capability index
Last model added
October 2026
Scores refreshed
November 20th, 2026
40
Entries on its own leaderboard, 32 models we track

Qwen3.8 Max (0902) leads with 1,462 Qwen3.8 Max (0902) leads with 1,462

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.

Bars start at 600.

  1. 1 Qwen3.8 Max (0902) 1,462
  2. 2 GLM-5.3 1,319
  3. 3 MiMo-V2.6 Pro 1,292
  4. 4 GLM-5.3-Flash 1,249
  5. 5 Qwen3.8 Flash 1,249
  6. 6 GLM-5.2 1,245
  7. 7 GLM-5.1 1,233
  8. 8 Kimi K3.1 Pro max 1,223
  9. 9 DeepSeek V4.1 Flash 1017 max 1,210
  10. 10 Kimi-K2.8-Preview 1,192
  11. 11 Kimi K3 1,155
  12. 12 DeepSeek V4.1 Flash max 1,151
  13. 13 MiMo-V2.6 Flash 1,107
  14. 14 DeepSeek-V4-Pro 1,100
  15. 15 Qwen3.7-Max 1,085
  16. 16 Qwen 3.6 Max (Preview) thinking 1,045
  17. 17 Claude Opus 4.5 reasoning 1,042
  18. 18 DeepSeek-V4-Flash(max) 1,038
  19. 19 MiniMax-M3 1,034
  20. 20 Kimi K2.6 thinking 1,027
  21. 21 Kimi K2.5 thinking 994
  22. 22 Kimi K2.7 Code 987
  23. 23 GLM-5 985
  24. 24 DeepSeek-V4-Flash(0731)(max) 981
  25. 25 Qwen3.6-Plus-2026-04-02-Thinking 978
  26. 26 Doubao-Seed-2-1-pro-260628(Thinking) 968
  27. 27 Gemini 3.5 Flash 921
  28. 28 GLM-4.7 903
  29. 29 DeepSeek-V3.2 862
  30. 30 GPT-5.2(high) 811
  31. 31 Gemini 3 Pro 809
  32. 32 Kimi K2 Thinking 794
  33. 33 Gemini 3 Flash 771
  34. 34 Qwen3.5 397B-A17B thinking 764
  35. 35 Qwen3-Max 743
  36. 36 MiniMax-M2.7 740
  37. 37 MiniMax-M2.1 731
  38. 38 Doubao-Seed-2.0-pro-260215(Thinking) 725
  39. 39 MiniMax-M2.5 720
  40. 40 Doubao-Seed-1.8-251228(Thinking) 670

Source: Official SuperCLUE Arena Web Coding leaderboard

Source: Official SuperCLUE Arena Web Coding leaderboard

#EntryScore ±SettingDateSource
1 Qwen3.8 Max (0902) 1,462 1420.4 to 1506.2 Sep 2026 Official leaderboard
2 GLM-5.3 1,319 1277.8 to 1365.3 Aug 2026 Official leaderboard
3 MiMo-V2.6 Pro 1,292 ±43.1 Sep 2026 Official leaderboard
4 GLM-5.3-Flash 1,249 ±43.8 Aug 2026 Official leaderboard
5 Qwen3.8 Flash 1,249 1203.6 to 1291.6 Aug 2026 Official leaderboard
6 GLM-5.2 1,245 ±27.3 Jun 2026 Official leaderboard
7 GLM-5.1 1,233 1209.1 to 1253.9 Apr 2026 Official leaderboard
8 Kimi K3.1 Pro 1,223 1148.4 to 1303.5 max Oct 2026 Official leaderboard
9 DeepSeek V4.1 Flash 1017 1,210 ±40.7 max Oct 2026 Official leaderboard
10 Kimi-K2.8-Preview 1,192 ±43.8 Official leaderboard
11 Kimi K3 1,155 ±35 Jul 2026 Official leaderboard
12 DeepSeek V4.1 Flash 1,151 ±40.6 max Sep 2026 Official leaderboard
13 MiMo-V2.6 Flash 1,107 ±43.3 Sep 2026 Official leaderboard
14 DeepSeek-V4-Pro 1,100 ±20.7 Apr 2026 Official leaderboard
15 Qwen3.7-Max 1,085 ±22.9 May 2026 Official leaderboard
16 Qwen 3.6 Max (Preview) 1,045 1024.4 to 1066.7 thinking Apr 2026 Official leaderboard
17 Claude Opus 4.5 1,042 1023.2 to 1058.3 reasoning Nov 2025 Official leaderboard
18 DeepSeek-V4-Flash(max) 1,038 1017.3 to 1060.2 Official leaderboard
19 MiniMax-M3 1,034 ±23.5 Jun 2026 Official leaderboard
20 Kimi K2.6 1,027 1003.4 to 1047.8 thinking Apr 2026 Official leaderboard
21 Kimi K2.5 994 975.8 to 1011.6 thinking Jan 2026 Official leaderboard
22 Kimi K2.7 Code 987 960.8 to 1011.7 Jun 2026 Official leaderboard
23 GLM-5 985 964.6 to 1003.8 Feb 2026 Official leaderboard
24 DeepSeek-V4-Flash(0731)(max) 981 947.7 to 1017.2 Official leaderboard
25 Qwen3.6-Plus-2026-04-02-Thinking 978 ±18.3 Official leaderboard
26 Doubao-Seed-2-1-pro-260628(Thinking) 968 ±26.9 Official leaderboard
27 Gemini 3.5 Flash 921 ±24.4 May 2026 Official leaderboard
28 GLM-4.7 903 ±18.1 Dec 2025 Official leaderboard
29 DeepSeek-V3.2 862 843.7 to 877.8 Dec 2025 Official leaderboard
30 GPT-5.2(high) 811 ±17.4 Official leaderboard
31 Gemini 3 Pro 809 ±16.9 Nov 2025 Official leaderboard
32 Kimi K2 Thinking 794 777.3 to 812.4 Nov 2025 Official leaderboard
33 Gemini 3 Flash 771 755.2 to 789.6 Dec 2025 Official leaderboard
34 Qwen3.5 397B-A17B 764 ±19.9 thinking Feb 2026 Official leaderboard
35 Qwen3-Max 743 ±16.9 Sep 2025 Official leaderboard
36 MiniMax-M2.7 740 ±20.4 Mar 2026 Official leaderboard
37 MiniMax-M2.1 731 ±17 Dec 2025 Official leaderboard
38 Doubao-Seed-2.0-pro-260215(Thinking) 725 706.2 to 745.3 Official leaderboard
39 MiniMax-M2.5 720 699.6 to 739.4 Feb 2026 Official leaderboard
40 Doubao-Seed-1.8-251228(Thinking) 670 ±18.3 Official leaderboard
#ModelScore ±SettingDateSource
1 Qwen3.8 Max (0902) 1,462 1420.4 to 1506.2 Sep 2026 Official leaderboard
2 GLM-5.3 1,319 1277.8 to 1365.3 Aug 2026 Official leaderboard
3 MiMo-V2.6 Pro 1,292 ±43.1 Sep 2026 Official leaderboard
4 GLM-5.3-Flash 1,249 ±43.8 Aug 2026 Official leaderboard
5 Qwen3.8 Flash 1,249 1203.6 to 1291.6 Aug 2026 Official leaderboard
6 GLM-5.2 1,245 ±27.3 Jun 2026 Official leaderboard
7 GLM-5.1 1,233 1209.1 to 1253.9 Apr 2026 Official leaderboard
8 Kimi K3.1 Pro 1,223 1148.4 to 1303.5 max Oct 2026 Official leaderboard
9 DeepSeek V4.1 Flash 1017 1,210 ±40.7 max Oct 2026 Official leaderboard
10 Kimi K3 1,155 ±35 Jul 2026 Official leaderboard
11 DeepSeek V4.1 Flash 1,151 ±40.6 max Sep 2026 Official leaderboard
12 MiMo-V2.6 Flash 1,107 ±43.3 Sep 2026 Official leaderboard
13 DeepSeek-V4-Pro 1,100 ±20.7 Apr 2026 Official leaderboard
14 Qwen3.7-Max 1,085 ±22.9 May 2026 Official leaderboard
15 Qwen 3.6 Max (Preview) 1,045 1024.4 to 1066.7 thinking Apr 2026 Official leaderboard
16 Claude Opus 4.5 1,042 1023.2 to 1058.3 reasoning Nov 2025 Official leaderboard
17 MiniMax-M3 1,034 ±23.5 Jun 2026 Official leaderboard
18 Kimi K2.6 1,027 1003.4 to 1047.8 thinking Apr 2026 Official leaderboard
19 Kimi K2.5 994 975.8 to 1011.6 thinking Jan 2026 Official leaderboard
20 Kimi K2.7 Code 987 960.8 to 1011.7 Jun 2026 Official leaderboard
21 GLM-5 985 964.6 to 1003.8 Feb 2026 Official leaderboard
22 Gemini 3.5 Flash 921 ±24.4 May 2026 Official leaderboard
23 GLM-4.7 903 ±18.1 Dec 2025 Official leaderboard
24 DeepSeek-V3.2 862 843.7 to 877.8 Dec 2025 Official leaderboard
25 Gemini 3 Pro 809 ±16.9 Nov 2025 Official leaderboard
26 Kimi K2 Thinking 794 777.3 to 812.4 Nov 2025 Official leaderboard
27 Gemini 3 Flash 771 755.2 to 789.6 Dec 2025 Official leaderboard
28 Qwen3.5 397B-A17B 764 ±19.9 thinking Feb 2026 Official leaderboard
29 Qwen3-Max 743 ±16.9 Sep 2025 Official leaderboard
30 MiniMax-M2.7 740 ±20.4 Mar 2026 Official leaderboard
31 MiniMax-M2.1 731 ±17 Dec 2025 Official leaderboard
32 MiniMax-M2.5 720 699.6 to 739.4 Feb 2026 Official leaderboard

8 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.

Qwen Moonshot AI Google DeepMind Anthropic Z.ai
Own leaderboard 8 leaders

Source: Official SuperCLUE Arena Web Coding 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
600 800 1,000 1,200 1,400 1,600 2026 2027 600 800 1,000 1,200 1,400 1,600 2026 2027

Source: Official SuperCLUE Arena Web Coding 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 Coding

All 523 in Coding
  • SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
  • LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests. DeepSeek-V4-Pro 93.5%
  • SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
  • DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
  • SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases. Seed2.1 Pro 59.8%
  • Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance. DeepSeek V4.1 Flash 3,471

Data audit

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (40 entries).
  • Pass Models scored 32 models we track, plus 8 other entries on its leaderboard.
  • Pass Newest score The newest is from October 2026, for Kimi K3.1 Pro.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Kimi K3.1 Pro (#17).
  • Note Headroom Scored as a rating, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on superclueai.com: 33 rows for models we track and 8 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