List your company

Agent Security Evaluation on Agent Security Bench (ASB) (DPI CA & CI Avg)

92.42ASRReAct
-0.587223.558947.70571.8511Jun 4, 2026

Evaluation results

MethodLinks
ReActBackbone=Gemini-2.5-Pro, Agent Category=Proprietary Agents2026.0692.426.016.26
ReActBackbone=Qwen3-32B, Agent Category=Open-Source Agents2026.0686.9607.35
ReActAgent Backbone=Average over four backbones (Qwen3-32B, Kimi-2.5, Gemini-2.5-Pro, GPT-5.1)2026.0679.9415.386.98
ReActBackbone=GPT-5.1, Agent Category=Proprietary Agents2026.0671.8636.34.51
ReActBackbone=Kimi-2.5, Agent Category=Open-Source Agents2026.0668.5319.199.8
TRIAD + Tri-GuardBackbone=GPT-5.1, Agent Category=Proprietary Agents2026.0617.472.387.5
TRIAD + TS-GuardBackbone=Kimi-2.5, Agent Category=Open-Source Agents2026.0613.2615.5160.15
TRIAD + Tri-GuardAgent Backbone=Average over four backbones (Qwen3-32B, Kimi-2.5, Gemini-2.5-Pro, GPT-5.1)2026.0613.0464.5224.73
TRIAD + TS-GuardBackbone=Gemini-2.5-Pro, Agent Category=Proprietary Agents2026.0612.674.0784.24
TRIAD + Tri-GuardBackbone=Kimi-2.5, Agent Category=Open-Source Agents2026.0612.0151.118.26
TRIAD + Tri-GuardBackbone=Qwen3-32B, Agent Category=Open-Source Agents2026.0611.5760.8332.94
TRIAD + Tri-GuardBackbone=Gemini-2.5-Pro, Agent Category=Proprietary Agents2026.0611.1873.7540.2
ToolSafeBackbone=Qwen3-32B, Agent Category=Open-Source Agents2026.0610.251.5444.98
TRIAD + TS-GuardBackbone=Qwen3-32B, Agent Category=Open-Source Agents2026.069.751.3388.8
TRIAD + TS-GuardBackbone=GPT-5.1, Agent Category=Proprietary Agents2026.068.418.1943.31
TRIAD + Qwen3.5-9BAgent Backbone=Average over four backbones (Qwen3-32B, Kimi-2.5, Gemini-2.5-Pro, GPT-5.1)2026.065.5626.360.66
ToolSafeBackbone=Gemini-2.5-Pro, Agent Category=Proprietary Agents2026.064.537.3374.19
ToolSafeBackbone=Kimi-2.5, Agent Category=Open-Source Agents2026.062.994.2457.21