SuperCLUE-WebCoding
Models implement web interfaces from specifications.
SuperCLUE reports scores in a public workbook.
- Made by
- SuperCLUE
- Kind
- Web and app front ends
- Level
- University
- Scored
- Checked automatically
- Answers
- Code
- Measures
- Capability
- Language
- Chinese
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
- Strongest model tested
- GPT-5.2 #51 on the capability index
- Last model added
- December 2025
- Scores refreshed
- November 20th, 2026
MiniMax-M2.1 leads with 75.7 among models released by December 2025 MiniMax-M2.1 leads with 75.7 among models released by December 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 December 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1
MiniMax-M2.1 API 75.7
- 2
Gemini 3 Flash API 72.6
- 3
Gemini 3 Pro API 72.5
- 4 Grok 4 API 72.2
- 5
Qwen3-Max API 71.9
- 6 Claude Opus 4.5 API 69.8
- 7 Kimi K2 Thinking API 68.9
- 8
Qwen3-Coder-480B-A35B API 67.9
- 9 GLM-4.6 API 67.9
- 10 GLM-4.7 API 66.9
- 11 GPT-5.2 API 65.3
- 12
DeepSeek-V3.2 API 64.4
- 13
Qwen3-235B-A22B-Thinking Jul 2025, API 60.8
- 14 Doubao-Seed-1.6-251015(Thinking) Thinking / API 39.9
Source: Official SuperCLUE-WebCoding leaderboard
- 1
MiniMax-M2.1 API 75.7
- 2
Gemini 3 Flash API 72.6
- 3
Gemini 3 Pro API 72.5
- 4 Grok 4 API 72.2
- 5
Qwen3-Max API 71.9
- 6 Claude Opus 4.5 API 69.8
- 7 Kimi K2 Thinking API 68.9
- 8
Qwen3-Coder-480B-A35B API 67.9
- 9 GLM-4.6 API 67.9
- 10 GLM-4.7 API 66.9
- 11 GPT-5.2 API 65.3
- 12
DeepSeek-V3.2 API 64.4
- 13
Qwen3-235B-A22B-Thinking Jul 2025, API 60.8
Source: Official SuperCLUE-WebCoding leaderboard
| # | Entry | Score | Setting | Released | Run on | Source |
|---|---|---|---|---|---|---|
| 1 | | 75.7 | API | Dec 2025 | Dec 2025 | Official leaderboard |
| 2 | | 72.6 | API | Dec 2025 | Dec 2025 | Official leaderboard |
| 3 | | 72.5 | API | Nov 2025 | Dec 2025 | Official leaderboard |
| 4 | Grok 4 | 72.2 | API | Jul 2025 | Dec 2025 | Official leaderboard |
| 5 | | 71.9 | API | Sep 2025 | Dec 2025 | Official leaderboard |
| 6 | Claude Opus 4.5 | 69.8 | API | Nov 2025 | Dec 2025 | Official leaderboard |
| 7 | Kimi K2 Thinking | 68.9 | API | Nov 2025 | Dec 2025 | Official leaderboard |
| 8 | | 67.9 | API | Jul 2025 | Dec 2025 | Official leaderboard |
| 9 | GLM-4.6 | 67.9 | API | Sep 2025 | Dec 2025 | Official leaderboard |
| 10 | GLM-4.7 | 66.9 | API | Dec 2025 | Dec 2025 | Official leaderboard |
| 11 | GPT-5.2 | 65.3 | API | Dec 2025 | Dec 2025 | Official leaderboard |
| 12 | | 64.4 | API | Dec 2025 | Dec 2025 | Official leaderboard |
| 13 | | 60.8 | API | Jul 2025 | Dec 2025 | Official leaderboard |
| 14 | Doubao-Seed-1.6-251015(Thinking) | 39.9 | Thinking / API | Dec 2025 | Official leaderboard |
| # | Model | Score | Setting | Released | Run on | Source |
|---|---|---|---|---|---|---|
| 1 | | 75.7 | API | Dec 2025 | Dec 2025 | Official leaderboard |
| 2 | | 72.6 | API | Dec 2025 | Dec 2025 | Official leaderboard |
| 3 | | 72.5 | API | Nov 2025 | Dec 2025 | Official leaderboard |
| 4 | Grok 4 | 72.2 | API | Jul 2025 | Dec 2025 | Official leaderboard |
| 5 | | 71.9 | API | Sep 2025 | Dec 2025 | Official leaderboard |
| 6 | Claude Opus 4.5 | 69.8 | API | Nov 2025 | Dec 2025 | Official leaderboard |
| 7 | Kimi K2 Thinking | 68.9 | API | Nov 2025 | Dec 2025 | Official leaderboard |
| 8 | | 67.9 | API | Jul 2025 | Dec 2025 | Official leaderboard |
| 9 | GLM-4.6 | 67.9 | API | Sep 2025 | Dec 2025 | Official leaderboard |
| 10 | GLM-4.7 | 66.9 | API | Dec 2025 | Dec 2025 | Official leaderboard |
| 11 | GPT-5.2 | 65.3 | API | Dec 2025 | Dec 2025 | Official leaderboard |
| 12 | | 64.4 | API | Dec 2025 | Dec 2025 | Official leaderboard |
| 13 | | 60.8 | API | Jul 2025 | Dec 2025 | Official leaderboard |
Rankings by part
We read 12 parts from the benchmark’s own leaderboard in 2 groups: 界面呈现榜单 and 交互功能榜单. Each ranks models on one part of the test. MiniMax-M2.1 leads 5 of them, and 4 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
界面呈现总分
12 models we track have a score here.
- 1
MiniMax-M2.1 89.8
- 2 Claude Opus 4.5 ↑4 86.0
- 3
Gemini 3 Flash ↓1 85.9
- 4
Qwen3-Max ↑1 85.4
- 5
Gemini 3 Pro ↓2 85.1
- 6 GPT-5.2 ↑5 82.8
- 7 GLM-4.6 ↑2 77.9
- 8
Qwen3-235B-A22B-Thinking Jul 2025 ↑5 77.8
- 9
DeepSeek-V3.2 ↑3 77.6
- 10 GLM-4.7 77.5
结构布局
12 models we track have a score here.
- 1 Grok 4 ↑3 90.5
- 2 GPT-5.2 ↑9 84.6
- 3
MiniMax-M2.1 ↓2 84.2
- 4
Qwen3-235B-A22B-Thinking Jul 2025 ↑9 80.6
- 5 GLM-4.7 ↑5 79.3
- 6
Gemini 3 Flash ↓4 78.8
- 7
Qwen3-Coder-480B-A35B ↑1 77.9
- 8
Gemini 3 Pro ↓5 77.5
- 9
Qwen3-Max ↓4 72.2
- 10 GLM-4.6 ↓1 72.2
视觉样式
12 models we track have a score here.
- 1 Claude Opus 4.5 ↑5 86.3
- 2
MiniMax-M2.1 ↓1 80.0
- 3 GLM-4.7 ↑7 77.9
- 4
Gemini 3 Pro ↓1 77.4
- 5
Qwen3-Max 76.1
- 6
DeepSeek-V3.2 ↑6 75.2
- 7 GPT-5.2 ↑4 75.1
- 8
Gemini 3 Flash ↓6 74.6
- 9 GLM-4.6 73.9
- 10
Qwen3-Coder-480B-A35B ↓2 71.5
信息内容呈现
12 models we track have a score here.
- 1
MiniMax-M2.1 98.5
- 2 Grok 4 ↑2 98.4
- 3 GPT-5.2 ↑8 98.2
- 4
Gemini 3 Pro ↓1 97.9
- 5
Gemini 3 Flash ↓3 96.8
- 6
Qwen3-235B-A22B-Thinking Jul 2025 ↑7 96.4
- 7 GLM-4.6 ↑2 95.4
- 8
Qwen3-Max ↓3 93.2
- 9
DeepSeek-V3.2 ↑3 92.5
- 10 Claude Opus 4.5 ↓4 92.4
响应式与适配
12 models we track have a score here.
- 1
Qwen3-Max ↑4 100.0
- 2
MiniMax-M2.1 ↓1 96.5
- 3 Claude Opus 4.5 ↑3 93.3
- 4
Gemini 3 Flash ↓2 93.3
- 5
Gemini 3 Pro ↓2 87.5
- 6
DeepSeek-V3.2 ↑6 73.3
- 7 GPT-5.2 ↑4 73.3
- 8
Qwen3-235B-A22B-Thinking Jul 2025 ↑5 70.6
- 9 GLM-4.6 70.0
- 10
Qwen3-Coder-480B-A35B ↓2 66.7
交互功能总分
12 models we track have a score here.
- 1
MiniMax-M2.1 73.6
- 2
Gemini 3 Flash 72.1
- 3 Grok 4 ↑1 70.9
- 4
Qwen3-Max ↑1 70.8
- 5
Gemini 3 Pro ↓2 70.0
- 6 GLM-4.6 ↑3 68.0
- 7
Qwen3-Coder-480B-A35B ↑1 67.7
- 8 Claude Opus 4.5 ↓2 66.4
- 9 GLM-4.7 ↑1 63.8
- 10 GPT-5.2 ↑1 63.3
导航视图切换
12 models we track have a score here.
- 1
Gemini 3 Flash ↑1 77.4
- 2
MiniMax-M2.1 ↓1 77.0
- 3 GLM-4.6 ↑6 76.1
- 4
Gemini 3 Pro ↓1 75.3
- 5
Qwen3-Max 72.3
- 6 Grok 4 ↓2 70.0
- 7 Claude Opus 4.5 ↓1 67.8
- 8 GLM-4.7 ↑2 62.0
- 9 GPT-5.2 ↑2 60.0
- 10
Qwen3-Coder-480B-A35B ↓2 59.1
输入与提交
12 models we track have a score here.
- 1
Gemini 3 Flash ↑1 97.3
- 2
Qwen3-Coder-480B-A35B ↑6 95.9
- 3
MiniMax-M2.1 ↓2 95.7
- 4 GPT-5.2 ↑7 94.8
- 5 Claude Opus 4.5 ↑1 94.4
- 6
Gemini 3 Pro ↓3 94.2
- 7
DeepSeek-V3.2 ↑5 94.0
- 8 GLM-4.6 ↑1 92.8
- 9 Grok 4 ↓5 89.0
- 10 GLM-4.7 88.7
数据列表管理
12 models we track have a score here.
- 1
Qwen3-Max ↑4 79.9
- 2
Qwen3-Coder-480B-A35B ↑6 73.7
- 3 Grok 4 ↑1 71.6
- 4 GPT-5.2 ↑7 70.5
- 5
Gemini 3 Flash ↓3 69.8
- 6
MiniMax-M2.1 ↓5 63.7
- 7 GLM-4.7 ↑3 62.2
- 8 GLM-4.6 ↑1 60.6
- 9
Gemini 3 Pro ↓6 52.1
- 10
DeepSeek-V3.2 ↑2 49.5
状态反馈提示
12 models we track have a score here.
- 1
MiniMax-M2.1 76.3
- 2 GLM-4.6 ↑7 73.1
- 3
Gemini 3 Flash ↓1 72.7
- 4
Gemini 3 Pro ↓1 71.6
- 5 Grok 4 ↓1 70.2
- 6
DeepSeek-V3.2 ↑6 69.0
- 7
Qwen3-Max ↓2 67.5
- 8 Claude Opus 4.5 ↓2 66.7
- 9 GPT-5.2 ↑2 66.0
- 10
Qwen3-235B-A22B-Thinking Jul 2025 ↑3 64.2
可视化统计
12 models we track have a score here.
- 1
MiniMax-M2.1 83.2
- 2 GPT-5.2 ↑9 82.9
- 3
Gemini 3 Flash ↓1 79.7
- 4
Gemini 3 Pro ↓1 79.3
- 5
Qwen3-Max 76.3
- 6 Grok 4 ↓2 76.2
- 7 GLM-4.7 ↑3 75.6
- 8
Qwen3-Coder-480B-A35B 74.8
- 9 GLM-4.6 74.2
- 10 Claude Opus 4.5 ↓4 70.1
多功能协同
12 models we track have a score here.
- 1 Claude Opus 4.5 ↑5 50.0
- 2 Grok 4 ↑2 48.0
- 3
Gemini 3 Pro 47.8
- 4
MiniMax-M2.1 ↓3 45.8
- 5 GLM-4.7 ↑5 43.9
- 6
Qwen3-Max ↓1 43.2
- 7
Qwen3-Coder-480B-A35B ↑1 39.3
- 8
DeepSeek-V3.2 ↑4 38.3
- 9
Gemini 3 Flash ↓7 35.6
- 10 GLM-4.6 ↓1 31.2
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.
Source: Official SuperCLUE-WebCoding leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official SuperCLUE-WebCoding 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
- 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
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 13 models we track, plus 1 other entry on its leaderboard.
- Warning Newest score The newest came 11 months ago, in December 2025, for MiniMax-M2.1.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.2 (#51).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on superclueai.com: 13 rows for models we track and 1 other entry 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.