List your company

ResearchFormerly WizWand

Probed Adversary Evaluation on 792 Privacy Evaluation Scenarios final benchmark 1.0

41HLSp ScoreKimi K2.5

Evaluation results

Method	Links
Kimi K2.52026.04		41	99	61
Gemini 3.1 Pro2026.04		40	99	61
MiniMax M2.12026.04		32	91	64
Kimi K2.52026.04		31	79	60
GLM-52026.04		30	93	68
GPT-5.42026.04		30	96	69
MiniMax M2.12026.04		30	90	66
Gemini 3.1 Pro2026.04		29	72	57
GPT-5.42026.04		27	85	65
MiniMax M2.12026.04		27	84	64
GLM-52026.04		25	79	64
GPT-5.42026.04		25	96	75
Kimi K2.52026.04		24	97	76
GLM-52026.04		23	93	75
Gemini 3.1 Pro2026.04		22	96	78
MiniMax M2.12026.04		22	82	68
GPT-5.42026.04		21	86	71
GLM-52026.04		18	81	71
Qwen3-8B2026.04		18	96	81
Kimi K2.52026.04		17	77	68
Gemini 3.1 Pro2026.04		16	82	73
MiniMax M2.52026.04		15	93	82
Qwen3-8B2026.04		15	94	83
Qwen3-8B2026.04		14	82	76
MiniMax M2.52026.04		13	83	78
MiniMax M2.52026.04		13	90	82
Qwen3-8B2026.04		12	85	79
MiniMax M2.52026.04		9	83	80