List your company

Agent Task Execution on QWENCLAWBENCH

59.5Overall ScoreClaude Opus 4.6
4.919.07533.2547.425Jun 15, 2026

Evaluation results

MethodLinks
Claude Opus 4.6Access=Closed-source2026.0659.557.857.355.648.554.376.867.371
GLM-5.1Access=Open-source / open-weight2026.0658.763.153.358.839.449.475.766.970.8
Qwen3.6-PlusAccess=Closed-source2026.0657.456.252.255.146.656.176.962.667.5
GPT-5.4Access=Closed-source2026.0656.755.254.357.938.452.670.671.163.7
MiMo-V2-ProAccess=Closed-source2026.0656.557.149.157.346.453.47460.664.9
Kimi-K2.5Access=Open-source / open-weight2026.0651.951.651.550.739.346.362.260.259.8
DeepSeek-V3.2-ThinkingAccess=Open-source / open-weight2026.0650.745.546.849.735.657.163.863.760.7
MiniMax-M2.7Access=Open-source / open-weight2026.0650.546.646.553.740.654.262.751.758.7
OpenClaw-Skill 9BBackbone=Qwen3.5-9B2026.0644.9132.343.360.214.228.470.978.450.1
Gemma4-31BAccess=Open-source / open-weight2026.0641.7————————
OpenClaw-Skill 4BBackbone=Qwen3.5-4B2026.0641.2935.846.146.77.833.251.56154.1
Gemma4-26B-A4BAccess=Open-source / open-weight2026.0638.7————————
Qwen3.5-9BAccess=Open-source / open-weight2026.0634.526.639.958.815.419.933.230.244.8
Qwen3.5-4BAccess=Open-source / open-weight2026.0631.524.328.834.215.434.348.355.424.4
OpenClaw-Skill-Qwen3 8BBackbone=Qwen3-8B2026.0615.8412.715.918.25.89.42234.313.3
OpenClaw-Skill-Qwen3 4BBackbone=Qwen3-4B2026.0612.8513.510.221.92.910.111.323.37.2
Qwen3-8BAccess=Open-source / open-weight2026.0611.517.37.913214.213.19.310.7
Qwen3-4BAccess=Open-source / open-weight2026.0676.610.87.10.87.22.56.313.2