Inference

QwenClawBench

Qwen's internal OpenClaw-style benchmark for measuring broad real-world agent performance across practical productivity and research tasks.

  • In index
  • Agentic
ModelsModels with a result.10
Top resultBest result on this test.64.3%Qwen3.7 Max
Top-3 spreadPoints from first to third.5.3 pts
YearYear of release.N/A

Result and price

020406080100$0.3$1$3$10$30RESULTOUTPUT PRICE PER 1M TOKENS · LOG SCALE

Results

10 results
#ModelResultThe score from the source.
01Qwen3.7 Max64.3%
02Qwen3.7 Plus61.8%
03Qwen 3.6 Max (preview)59.0%
04Qwen3.6 Plus57.2%
05Kimi K2.554.3%
06GLM-554.1%
07Qwen3.6-27B53.4%
08Qwen3.6-35B-A3B52.6%
09Claude Opus 4.552.3%
10Qwen3.5 397B51.8%