List your company

Reasoning length evaluation on 20 attack prompts

7,745Avg LengthRECUR
-78.921,952.293,983.56,014.71Feb 9, 2026

Evaluation results

MethodLinks
RECURModel=QWQ-32B2026.027,74515,537
RECURModel=O1-HIGH2026.026,55221,549
RECURModel=DS-QWEN-14B2026.026,08914,834
RECURModel=DS-LLAMA-8B2026.025,88415,674
RECURModel=DS-V3.22026.025,30610,999
AUTODOSModel=QWQ-32B2026.024,83311,329
RECURModel=GEMINI-2.52026.024,46164,440
OVERTHINKModel=O1-HIGH2026.023,94213,824
GCGModel=DS-V3.22026.023,77110,933
OVERTHINKModel=GEMINI-2.52026.023,73519,735
GCGModel=QWQ-32B2026.023,44516,301
RECURModel=O1-MEDIUM2026.023,4349,565
RECURModel=GROK-4-FAST2026.023,2268,639
GCGModel=DS-QWEN-14B2026.023,14616,305
LOOPLLMModel=QWQ-32B2026.022,64016,334
OVERTHINKModel=GROK-4-FAST2026.022,2715,186
AUTODOSModel=DS-LLAMA-8B2026.022,0484,470
LOOPLLMModel=DS-LLAMA-8B2026.021,94416,326
AUTODOSModel=DS-QWEN-14B2026.021,8855,366
GCGModel=O1-HIGH2026.021,8012,688
BASELINEModel=QWQ-32B2026.021,7964,459
AUTODOSModel=GEMINI-2.52026.021,3202,493
OVERTHINKModel=QWQ-32B2026.021,2771,688
OVERTHINKModel=O1-MEDIUM2026.021,2561,984
GCGModel=DS-LLAMA-8B2026.021,2113,407
AUTODOSModel=O1-HIGH2026.021,1081,597
GCGModel=O1-MEDIUM2026.029601,984
LOOPLLMModel=GEMINI-2.52026.028671,296
LOOPLLMModel=O1-HIGH2026.028382,944
BASELINEModel=GEMINI-2.52026.027522,264
LOOPLLMModel=O1-MEDIUM2026.027041,408
LOOPLLMModel=DS-QWEN-14B2026.026981,981
LOOPLLMModel=DS-V3.22026.026892,833
BASELINEModel=O1-HIGH2026.026493,089
OVERTHINKModel=DS-LLAMA-8B2026.026291,003
BASELINEModel=DS-LLAMA-8B2026.026162,298
OVERTHINKModel=DS-V3.22026.02594939
AUTODOSModel=O1-MEDIUM2026.025821,337
OVERTHINKModel=DS-QWEN-14B2026.02580840
AUTODOSModel=DS-V3.22026.025581,554
BASELINEModel=DS-V3.22026.025531,923
LOOPLLMModel=GROK-4-FAST2026.025301,640
BASELINEModel=DS-QWEN-14B2026.025211,012
BASELINEModel=O1-MEDIUM2026.02425970
BASELINEModel=GROK-4-FAST2026.024172,136
GCGModel=GROK-4-FAST2026.02371481
AUTODOSModel=GROK-4-FAST2026.02332555
GCGModel=GEMINI-2.52026.02222861