List your company

Safety-Utility Trade-off Evaluation on AgentHarm

82.79HS ScoreTRIAD + Tri-Guard
34.419646.977359.53572.0927Jun 4, 2026

Evaluation results

MethodLinks
TRIAD + Tri-GuardBackbone=Kimi-2.5, Agent Category=Open-Source Agents2026.0682.7913.26
TRIAD + TS-GuardBackbone=Kimi-2.5, Agent Category=Open-Source Agents2026.0682.2515.32
TRIAD + Tri-GuardBackbone=Gemini-2.5-Pro, Agent Category=Proprietary Agents2026.0680.8515.39
TRIAD + Tri-GuardBackbone=Qwen3-32B, Agent Category=Open-Source Agents2026.0680.329.87
ReActBackbone=GPT-5.1, Agent Category=Proprietary Agents2026.0680.3121.71
TRIAD + Tri-GuardBackbone=GPT-5.1, Agent Category=Proprietary Agents2026.0679.7313.66
TRIAD + TS-GuardBackbone=GPT-5.1, Agent Category=Proprietary Agents2026.0677.5815.08
TRIAD + TS-GuardBackbone=Gemini-2.5-Pro, Agent Category=Proprietary Agents2026.0676.7214.81
TRIAD + TS-GuardBackbone=Qwen3-32B, Agent Category=Open-Source Agents2026.0676.158.89
ReActBackbone=Kimi-2.5, Agent Category=Open-Source Agents2026.0674.5736.9
ToolSafeBackbone=Kimi-2.5, Agent Category=Open-Source Agents2026.0668.716.59
ToolSafeBackbone=Gemini-2.5-Pro, Agent Category=Proprietary Agents2026.0668.4910.38
ToolSafeBackbone=Qwen3-32B, Agent Category=Open-Source Agents2026.0667.3111.57
ReActBackbone=Gemini-2.5-Pro, Agent Category=Proprietary Agents2026.0658.5455.67
ReActBackbone=Qwen3-32B, Agent Category=Open-Source Agents2026.0636.2877.04