List your company

Safety Evaluation on Fortress (JailBreak Score)

2.8JailBreak ScoreSInternal
1.9047.9521420.048May 9, 2026

Evaluation results

MethodLinks
SInternalModel Backbone=DeepSeek-R1-Distill-Llama-8B, RL Training (GRPO)=true2026.052.8
STAR-1Model Backbone=DeepSeek-R1-Distill-Llama-8B, RL Training (GRPO)=true2026.054.8
SInternalModel Backbone=DeepSeek-R1-Distill-Qwen-14B, RL Training (GRPO)=true2026.055.2
STAR-1Model Backbone=DeepSeek-R1-Distill-Qwen-14B, RL Training (GRPO)=true2026.057.8
SafeChainModel Backbone=DeepSeek-R1-Distill-Qwen-14B, RL Training (GRPO)=true2026.058
SInternalModel Backbone=DeepSeek-R1-Distill-Qwen-7B, RL Training (GRPO)=true2026.0511
BaseModel Backbone=DeepSeek-R1-Distill-Qwen-14B, RL Training (GRPO)=true2026.0512.2
SafeChainModel Backbone=DeepSeek-R1-Distill-Llama-8B, RL Training (GRPO)=true2026.0514.4
BaseModel Backbone=DeepSeek-R1-Distill-Llama-8B, RL Training (GRPO)=true2026.0514.8
STAR-1Model Backbone=DeepSeek-R1-Distill-Qwen-7B, RL Training (GRPO)=true2026.0517
BaseModel Backbone=DeepSeek-R1-Distill-Qwen-7B, RL Training (GRPO)=true2026.0521.4
SafeChainModel Backbone=DeepSeek-R1-Distill-Qwen-7B, RL Training (GRPO)=true2026.0525.2