List your company

Jailbreak Attack on StrongREJECT (test)

82.34ScoreGCD
-3.293618.938241.1763.4018Jun 14, 2026

Evaluation results

MethodLinks
GCDVictim Model=MISTRAL-7B, Perplexity Defense=Disabled2026.0682.34————
GCDVictim Model=MISTRAL-7B, Perplexity Defense=Enabled2026.0682.34————
AUTODANVictim Model=MISTRAL-7B, Perplexity Defense=Disabled2026.0681.4————
AUTODANVictim Model=MISTRAL-7B, Perplexity Defense=Enabled2026.0681.4————
AUTODANVictim Model=QWEN-2.5-7B, Perplexity Defense=Disabled2026.0680.54————
AUTODANVictim Model=QWEN-2.5-7B, Perplexity Defense=Enabled2026.0680.54————
GCDVictim Model=QWEN-2.5-7B, Perplexity Defense=Disabled2026.0677.89————
GCDVictim Model=QWEN-2.5-7B, Perplexity Defense=Enabled2026.0677.89————
INPAINTINGVictim Model=MISTRAL-7B, Perplexity Defense=Disabled2026.0673.81————
INPAINTINGVictim Model=MISTRAL-7B, Perplexity Defense=Enabled2026.0673.81————
GCDVictim Model=LLAMA-3-8B, Perplexity Defense=Disabled2026.0671.59————
GCDVictim Model=LLAMA-3-8B, Perplexity Defense=Enabled2026.0671.59————
GCDVictim Model=LLAMA-3-70B, Perplexity Defense=Disabled2026.0665.71————
GCDVictim Model=LLAMA-3-70B, Perplexity Defense=Enabled2026.0665.71————
GCGVictim Model=MISTRAL-7B, Perplexity Defense=Disabled2026.0664.4————
GBDAVictim Model=MISTRAL-7B, Perplexity Defense=Disabled2026.0656.3————
Direct RequestVictim Model=MISTRAL-7B, Perplexity Defense=Disabled2026.0653.2————
PAIRVictim Model=QWEN-2.5-7B, Perplexity Defense=Disabled2026.0651.57————
PAIRVictim Model=QWEN-2.5-7B, Perplexity Defense=Enabled2026.0651.57————
TAPVictim Model=QWEN-2.5-7B, Perplexity Defense=Disabled2026.0650.84————
TAPVictim Model=QWEN-2.5-7B, Perplexity Defense=Enabled2026.0650.84————
Direct RequestVictim Model=MISTRAL-7B, Perplexity Defense=Enabled2026.0650.82————
PAIRVictim Model=MISTRAL-7B, Perplexity Defense=Disabled2026.0650.05————
PAIRVictim Model=MISTRAL-7B, Perplexity Defense=Enabled2026.0650.05————
TAPVictim Model=MISTRAL-7B, Perplexity Defense=Disabled2026.0649.09————
TAPVictim Model=MISTRAL-7B, Perplexity Defense=Enabled2026.0649.09————
GCGVictim Model=QWEN-2.5-7B, Perplexity Defense=Disabled2026.0648.54————
INPAINTINGVictim Model=QWEN-2.5-7B, Perplexity Defense=Disabled2026.0643.06————
INPAINTINGVictim Model=QWEN-2.5-7B, Perplexity Defense=Enabled2026.0643.06————
PAIRVictim Model=LLAMA-3-70B, Perplexity Defense=Disabled2026.0630.33————
PAIRVictim Model=LLAMA-3-70B, Perplexity Defense=Enabled2026.0630.33————
INPAINTINGVictim Model=LLAMA-3-70B, Perplexity Defense=Disabled2026.0626.38————
INPAINTINGVictim Model=LLAMA-3-70B, Perplexity Defense=Enabled2026.0626.38————
INPAINTINGVictim Model=LLAMA-3-8B, Perplexity Defense=Disabled2026.0615.75————
INPAINTINGVictim Model=LLAMA-3-8B, Perplexity Defense=Enabled2026.0615.75————
AUTODANVictim Model=LLAMA-3-8B, Perplexity Defense=Disabled2026.0615.68————
AUTODANVictim Model=LLAMA-3-8B, Perplexity Defense=Enabled2026.0615.68————
PAIRVictim Model=LLAMA-3-8B, Perplexity Defense=Disabled2026.0614.36————
PAIRVictim Model=LLAMA-3-8B, Perplexity Defense=Enabled2026.0614.36————
TAPVictim Model=LLAMA-3-70B, Perplexity Defense=Disabled2026.0612.94————
TAPVictim Model=LLAMA-3-70B, Perplexity Defense=Enabled2026.0612.94————
Direct RequestVictim Model=QWEN-2.5-7B, Perplexity Defense=Disabled2026.0611.99————
Direct RequestVictim Model=QWEN-2.5-7B, Perplexity Defense=Enabled2026.0611.99————
GBDAVictim Model=QWEN-2.5-7B, Perplexity Defense=Disabled2026.069.34————
TAPVictim Model=LLAMA-3-8B, Perplexity Defense=Disabled2026.065.53————
TAPVictim Model=LLAMA-3-8B, Perplexity Defense=Enabled2026.065.53————
Direct RequestVictim Model=LLAMA-3-70B, Perplexity Defense=Disabled2026.065.33————
Direct RequestVictim Model=LLAMA-3-70B, Perplexity Defense=Enabled2026.065.33————
GCGVictim Model=LLAMA-3-70B, Perplexity Defense=Disabled2026.063.69————
Direct RequestVictim Model=LLAMA-3-8B, Perplexity Defense=Disabled2026.063.68————
Direct RequestVictim Model=LLAMA-3-8B, Perplexity Defense=Enabled2026.063.68————
GBDAVictim Model=LLAMA-3-70B, Perplexity Defense=Disabled2026.063.54————
GCGVictim Model=LLAMA-3-8B, Perplexity Defense=Disabled2026.061.04————
GBDAVictim Model=LLAMA-3-8B, Perplexity Defense=Disabled2026.060.39————
AUTODANVictim Model=LLAMA-3-70B, Perplexity Defense=Disabled2026.060.37————
AUTODANVictim Model=LLAMA-3-70B, Perplexity Defense=Enabled2026.060.37————
GCGVictim Model=LLAMA-3-8B, Perplexity Defense=Enabled2026.060————
GBDAVictim Model=LLAMA-3-8B, Perplexity Defense=Enabled2026.060————
GCGVictim Model=QWEN-2.5-7B, Perplexity Defense=Enabled2026.060————
GBDAVictim Model=QWEN-2.5-7B, Perplexity Defense=Enabled2026.060————
GCGVictim Model=MISTRAL-7B, Perplexity Defense=Enabled2026.060————
GBDAVictim Model=MISTRAL-7B, Perplexity Defense=Enabled2026.060————
GCGVictim Model=LLAMA-3-70B, Perplexity Defense=Enabled2026.060————
GBDAVictim Model=LLAMA-3-70B, Perplexity Defense=Enabled2026.060————
CKA-AgentAttack model=Qwen3-32B-abliterated (Thinking), Judge model=Gemini-2.5-Pro, Method type=Decomposition2025.12—98.80.600.6
CKA-AgentAttack model=Qwen3-32B-abliterated (Thinking), Judge model=GPT-5-mini, Method type=Decomposition2025.12—97.52.500
CKA-AgentAttack model=Qwen3-32B-abliterated (Thinking), Judge model=Fine-Tuned Judge (Gemma-2b), Method type=Decomposition2025.12—98.81.200