Anthropic Claude 3.5 Sonnet v1 vs OpenAI GPT-5.2
Detailed comparison for LLMs
AnthropicOpenAI
Head-to-Head Overview
GPT-5.2 is the overall winner in this comparison!
LLM Metric
Comparison of Attack Success Rate (ASR) metrics for Anthropic Claude 3.5 Sonnet v1 and OpenAI GPT-5.2 across the three attack methods.
Overall (ASR)
Claude 3.5 Sonnet v1
16.010
GPT-5.2
0
TAP Attack Method (ASR)
Claude 3.5 Sonnet v1
28.390
GPT-5.2
0
Crescendo Attack Method (ASR)
Claude 3.5 Sonnet v1
19.310
GPT-5.2
0
Zero-Shot (ASR)
Claude 3.5 Sonnet v1
0.350
GPT-5.2
0
Key Highlights
- OpenAI GPT-5.2 has a lower Overall (ASR).
- OpenAI GPT-5.2 has a lower TAP Attack Method (ASR).
- OpenAI GPT-5.2 has a lower Crescendo Attack Method (ASR).
- OpenAI GPT-5.2 has a lower Zero-Shot (ASR).
Metric Radar Chart
claude-3-5-sonnet-v1
gpt-5-2