List your company

Jailbreak Attack on JailbreakBench (JBB)

0ASRCoA
-3.80621.884547.57573.2655Dec 8, 2025Dec 28, 2025Jan 17, 2026Feb 7, 2026Feb 27, 2026Mar 19, 2026Apr 9, 2026

Evaluation results

MethodLinks
CoAVictim LLM=Gemma-2-9B-IT, Turn Type=Multi-Turn2025.120———
NoneModel=Gemma 2 2B2026.040———
CoAVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Multi-Turn2025.121.82———
NoneModel=Llama 3.2 3B2026.042———
CoAVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Multi-Turn2025.123.64———
CoAVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Multi-Turn2025.1218.18———
ArtPromptVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Single-Turn2025.1227.27———
ArtPromptVictim LLM=Gemma-2-9B-IT, Turn Type=Single-Turn2025.1229.09———
Jailbreak-R1Victim LLM=Gemma-2-9B-IT, Turn Type=Single-Turn2025.1232.12———
Jailbreak-R1Victim LLM=Llama-3.1-8B-Instruct, Turn Type=Single-Turn2025.1240———
DeepInception (Text Baseline)Target Model=GPT-4o Realtime, Modality=Text2026.0140———
ArtPromptVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Single-Turn2025.1241.82———
SirenVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Multi-Turn2025.1242.42———
SirenVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Multi-Turn2025.1243.03———
MTSAVictim LLM=Gemma-2-9B-IT, Turn Type=Multi-Turn2025.1252.73———
X-TeamingVictim LLM=Gemma-2-9B-IT, Turn Type=Multi-Turn2025.1252.73———
ReNeLLMVictim LLM=Gemma-2-9B-IT, Turn Type=Single-Turn2025.1254.55———
SirenVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Multi-Turn2025.1254.55———
AutoDan-TurboVictim LLM=Gemma-2-9B-IT, Turn Type=Single-Turn2025.1255.76———
ActorAttackVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Multi-Turn2025.1256.36———
SirenVictim LLM=Gemma-2-9B-IT, Turn Type=Multi-Turn2025.1259.39———
ActorAttackVictim LLM=Gemma-2-9B-IT, Turn Type=Multi-Turn2025.1260———
MTSAVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Multi-Turn2025.1260———
AutoDan-TurboVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Single-Turn2025.1260.61———
ArtPromptVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Single-Turn2025.1261.82———
Jailbreak-R1Victim LLM=Qwen2.5-7B-Instruct, Turn Type=Single-Turn2025.1261.82———
GRPOVictim LLM=Gemma-2-9B-IT, Turn Type=Multi-Turn2025.1263.03———
AutoDan-TurboVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Single-Turn2025.1263.64———
AutoDan-TurboVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Single-Turn2025.1263.64———
ReNeLLMVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Single-Turn2025.1265.45———
Audio Narrative AttacksTarget Model=GPT-4o Realtime, Modality=Audio2026.0166.67———
Audio Narrative AttacksTarget Model=Qwen2.5-Omni, Modality=Audio2026.0170.33———
X-TeamingVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Multi-Turn2025.1270.91———
TROJailVictim LLM=Gemma-2-9B-IT, Turn Type=Multi-Turn2025.1272.12———
ActorAttackVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Multi-Turn2025.1272.73———
GRPO w/ IPRVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Multi-Turn2025.1273.33———
Jailbreak-R1Victim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Single-Turn2025.1273.94———
ActorAttackVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Multi-Turn2025.1274.55———
GRPOVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Multi-Turn2025.1275.76———
DeepInception (Text Baseline)Target Model=Qwen2.5-Omni, Modality=Text2026.0176.67———
TROJailVictim LLM=Llama-3.1-8B-Instruct, Turn Type=Multi-Turn2025.1277.58———
ReNeLLMVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Single-Turn2025.1280———
MTSAVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Multi-Turn2025.1280———
DIMModel=Gemma 2 2B2026.0480———
POModel=Gemma 2 2B2026.0480———
POModel=Llama 3.2 3B2026.0481———
ReNeLLMVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Single-Turn2025.1281.82———
NTPModel=Llama 3.2 3B2026.0483———
GRPO w/ IPRVictim LLM=Gemma-2-9B-IT, Turn Type=Multi-Turn2025.1283.03———
X-TeamingVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Multi-Turn2025.1283.64———
DeepInception (Text Baseline)Target Model=Gemini 2.0 Flash, Modality=Text2026.0184———
NTPModel=Gemma 2 2B2026.0484———
DIMModel=Llama 3.2 3B2026.0485———
GRPOVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Multi-Turn2025.1285.45———
MTSAVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Multi-Turn2025.1287.27———
Audio Narrative AttacksTarget Model=Gemini 2.0 Flash, Modality=Audio2026.0188.33———
GRPOVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Multi-Turn2025.1288.48———
X-TeamingVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Multi-Turn2025.1289.09———
TROJailVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Multi-Turn2025.1290.91———
GRPO w/ IPRVictim LLM=Qwen2.5-7B-Instruct, Turn Type=Multi-Turn2025.1293.33———
GRPO w/ IPRVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Multi-Turn2025.1293.94———
TROJailVictim LLM=Mistral-7B-Instruct-v0.3, Turn Type=Multi-Turn2025.1295.15———
ArtPromptTarget Model=DeepSeek-V32026.01—27817
ArtPromptTarget Model=Qwen3-32B2026.01—201830
ArtPromptTarget Model=GPT-OSS-120B2026.01—2795
BaselineTarget Model=DeepSeek-V32026.01—14759
BaselineTarget Model=Qwen3-32B2026.01—10965
BaselineTarget Model=GPT-OSS-120B2026.01—1199
DRATarget Model=DeepSeek-V32026.01—224338
DRATarget Model=Qwen3-32B2026.01—211213
DRATarget Model=GPT-OSS-120B2026.01—61485
FlipAttackTarget Model=DeepSeek-V32026.01—26100
FlipAttackTarget Model=Qwen3-32B2026.01—163428
FlipAttackTarget Model=GPT-OSS-120B2026.01—1199
iMISTTarget Model=DeepSeek-V32026.01—50644
iMISTTarget Model=Qwen3-32B2026.01—46540
iMISTTarget Model=GPT-OSS-120B2026.01—383036
PAIRTarget Model=DeepSeek-V32026.01—304510
PAIRTarget Model=Qwen3-32B2026.01—362415
PAIRTarget Model=GPT-OSS-120B2026.01—3692
PASSTarget Model=DeepSeek-V32026.01—28569
PASSTarget Model=Qwen3-32B2026.01—232015
PASSTarget Model=GPT-OSS-120B2026.01—81672
RL-JackTarget Model=DeepSeek-V32026.01—183427
RL-JackTarget Model=Qwen3-32B2026.01—171238
RL-JackTarget Model=GPT-OSS-120B2026.01—00100
SATATarget Model=DeepSeek-V32026.01—16644
SATATarget Model=Qwen3-32B2026.01—152247
SATATarget Model=GPT-OSS-120B2026.01—2890
TAPTarget Model=DeepSeek-V32026.01—11819
TAPTarget Model=Qwen3-32B2026.01—404212
TAPTarget Model=GPT-OSS-120B2026.01—2494
WordGameTarget Model=DeepSeek-V32026.01—13949
WordGameTarget Model=Qwen3-32B2026.01—244022
WordGameTarget Model=GPT-OSS-120B2026.01—121465