reasoning benchmark
QwenWorldBench
QwenWorldBench is Qwen's internal benchmark for evaluating LLMs as world models that simulate agentic environments across Terminal, SWE, MCP, Search, OS, Android, and Web domains.
Updated Aug 7, 2026
Published models2
Registry coverage2
MetricScore
EvidenceB
QwenWorldBench leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 62.1% | 100.0% | 2 | C | |
| 2 | AC | 57.3% | 0.0% | 2 | C |
Score distribution
Top published rows on the benchmark's original scale.
QwenWorldBench highlights
The top published results on this benchmark's own scale.
What is QwenWorldBench?
Definition and scoring fields from the benchmark registry.
QwenWorldBench is Qwen's internal benchmark for evaluating LLMs as world models that simulate agentic environments across Terminal, SWE, MCP, Search, OS, Android, and Web domains.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- QwenWorldBench
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- qwenworldbench|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about QwenWorldBench.
Which model scores highest on QwenWorldBench?
Qwen3.7-Plus is currently ranked first with 62.1%.
What does QwenWorldBench measure?
QwenWorldBench is Qwen's internal benchmark for evaluating LLMs as world models that simulate agentic environments across Terminal, SWE, MCP, Search, OS, Android, and Web domains.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.