SuperCLUE-OpenClaw
Models complete coding, memory, and writing tasks in the OpenClaw environment.
SuperCLUE reports scores in a public workbook.
- Made by
- SuperCLUE
- Kind
- General assistants
- Level
- University
- Scored
- Settled by events
- Answers
- Agent task
- Measures
- Capability
- Language
- Chinese
- Interaction
- Agentic
- Environment
- OpenClaw
- Refresh
- One fixed set
- Run by
- A host
- In views
- Agents
- Strongest model tested
- GPT-5.4 #28 on the capability index
- Last model added
- March 2026
- Scores refreshed
- November 20th, 2026
Claude Opus 4.6 leads with 92.3 among models released by March 2026 Claude Opus 4.6 leads with 92.3 among models released by March 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 March 2026, 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 Claude Opus 4.6 API 92.3
- 2 GPT-5.4 API 92.2
- 3 Doubao-Seed-2.0-pro-260215 API 92.0
- 4
MiniMax-M2.5 API 87.2
- 5
Qwen3-Max API 87.1
- 6 Kimi K2.5 API 86.9
- 7 GLM-5 API 86.3
- 8
Gemini 3.1 Pro API 80.0
- 9
DeepSeek-V3.2 API 61.0
Source: Official SuperCLUE-OpenClaw leaderboard
- 1 Claude Opus 4.6 API 92.3
- 2 GPT-5.4 API 92.2
- 3
MiniMax-M2.5 API 87.2
- 4
Qwen3-Max API 87.1
- 5 Kimi K2.5 API 86.9
- 6 GLM-5 API 86.3
- 7
Gemini 3.1 Pro API 80.0
- 8
DeepSeek-V3.2 API 61.0
Source: Official SuperCLUE-OpenClaw leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 4.6 | 92.3 | API | Feb 2026 | Official leaderboard |
| 2 | GPT-5.4 | 92.2 | API | Mar 2026 | Official leaderboard |
| 3 | Doubao-Seed-2.0-pro-260215 closed weights | 92.0 | API | Official leaderboard | |
| 4 | | 87.2 | API | Feb 2026 | Official leaderboard |
| 5 | | 87.1 | API | Sep 2025 | Official leaderboard |
| 6 | Kimi K2.5 | 86.9 | API | Jan 2026 | Official leaderboard |
| 7 | GLM-5 | 86.3 | API | Feb 2026 | Official leaderboard |
| 8 | | 80.0 | API | Feb 2026 | Official leaderboard |
| 9 | | 61.0 | API | Dec 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 4.6 | 92.3 | API | Feb 2026 | Official leaderboard |
| 2 | GPT-5.4 | 92.2 | API | Mar 2026 | Official leaderboard |
| 3 | | 87.2 | API | Feb 2026 | Official leaderboard |
| 4 | | 87.1 | API | Sep 2025 | Official leaderboard |
| 5 | Kimi K2.5 | 86.9 | API | Jan 2026 | Official leaderboard |
| 6 | GLM-5 | 86.3 | API | Feb 2026 | Official leaderboard |
| 7 | | 80.0 | API | Feb 2026 | Official leaderboard |
| 8 | | 61.0 | API | Dec 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 编码能力, 记忆能力 and 内容创作. Claude Opus 4.6 leads 3 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.
编码能力
8 models we track have a score here.
- 1 Claude Opus 4.6 98.6
- 2
Gemini 3.1 Pro ↑5 98.6
- 3
Qwen3-Max ↑1 97.1
- 4 Kimi K2.5 ↑1 97.1
- 5 GLM-5 ↑1 97.1
- 6 GPT-5.4 ↓4 97.1
- 7
MiniMax-M2.5 ↓4 94.6
- 8
DeepSeek-V3.2 90.5
记忆能力
8 models we track have a score here.
- 1
Qwen3-Max ↑3 81.3
- 2 Kimi K2.5 ↑3 81.3
- 3 GLM-5 ↑3 81.3
- 4
MiniMax-M2.5 ↓1 81.3
- 5 GPT-5.4 ↓3 80.8
- 6 Claude Opus 4.6 ↓5 79.2
- 7
DeepSeek-V3.2 ↑1 61.3
- 8
Gemini 3.1 Pro ↓1 60.8
内容创作
8 models we track have a score here.
- 1 Claude Opus 4.6 89.2
- 2 GPT-5.4 85.5
- 3 GLM-5 ↑3 83.3
- 4 Kimi K2.5 ↑1 82.3
- 5
MiniMax-M2.5 ↓2 80.8
- 6
Qwen3-Max ↓2 75.6
- 7
Gemini 3.1 Pro 75.2
- 8
DeepSeek-V3.2 50.0
数据处理
8 models we track have a score here.
- 1 GPT-5.4 ↑1 97.9
- 2 Claude Opus 4.6 ↓1 95.8
- 3 GLM-5 ↑3 95.1
- 4
MiniMax-M2.5 ↓1 92.3
- 5
Qwen3-Max ↓1 89.5
- 6 Kimi K2.5 ↓1 70.8
- 7
Gemini 3.1 Pro 67.3
- 8
DeepSeek-V3.2 47.9
研究分析
8 models we track have a score here.
- 1 Kimi K2.5 ↑4 95.7
- 2 GPT-5.4 94.8
- 3 Claude Opus 4.6 ↓2 92.2
- 4
Qwen3-Max 90.6
- 5
MiniMax-M2.5 ↓2 85.8
- 6
Gemini 3.1 Pro ↑1 84.4
- 7 GLM-5 ↓1 75.9
- 8
DeepSeek-V3.2 56.0
平均耗时
8 models we track have a score here.
- 1
DeepSeek-V3.2 ↑7 161.1
- 2 GLM-5 ↑4 94.9
- 3
MiniMax-M2.5 94.2
- 4
Qwen3-Max 69.7
- 5 Claude Opus 4.6 ↓4 68.1
- 6 Kimi K2.5 ↓1 57.6
- 7 GPT-5.4 ↓5 41.7
- 8
Gemini 3.1 Pro ↓1 39.1
成本(元/百万Tokens)
8 models we track have a score here.
- 1 Claude Opus 4.6 69.1
- 2 GPT-5.4 39.4
- 3
Gemini 3.1 Pro ↑4 30.9
- 4 Kimi K2.5 ↑1 8.25
- 5 GLM-5 ↑1 7.50
- 6
Qwen3-Max ↓2 4.38
- 7
MiniMax-M2.5 ↓4 3.68
- 8
DeepSeek-V3.2 2.25
2 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-OpenClaw 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-OpenClaw 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 Tool use
- AutomationBench In use Agent completes end-to-end business workflows inside a simulated company exposing dozens of REST APIs. Graded on resulting state.
Gemini 4 Neon 56.3%
- MCP Atlas In use Multi-step workflows over real MCP servers. Scores tool-use competency on realistic tasks.
Gemini 4 Argon 90.7%
- BFCL V4 In use Models call functions and tools, including multi-turn agentic tasks with web search and memory. Checked against expected calls and outcomes.
Qwen3.7-Max 75%
- τ³-bench Banking In use Agents act as banking customer-service reps over many turns with a simulated user, following policy. Pass^1 on the final database state.
Qwen 3.8 Max 55.1%
- Toolathlon-Verified In use 108 long-horizon tasks across 32 apps through about 600 MCP tools. Execution-based checkers inspect final environment state. GLM-5.3-Flash 78.4%
- DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%
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 (9 entries).
- Pass Models scored 8 models we track, plus 1 other entry on its leaderboard.
- Warning Newest score The newest came 8 months ago, in March 2026, for GPT-5.4.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.4 (#28).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on superclueai.com: 8 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.