Detailed comparison for LLMs
It's a close match between DeepSeek-V3.2 Non-thinking and Llama 3.1 Nemotron Ultra 253B v1!
Comparison of Attack Success Rate (ASR) metrics for DeepSeek DeepSeek-V3.2 Non-thinking and NVIDIA Llama 3.1 Nemotron Ultra 253B v1 across the three attack methods.
No specific highlights generated for this comparison.