Alibaba Qwen 2.5 72B vs Anthropic Claude 3.5 Sonnet v1
Detailed comparison for LLMs
AlibabaAnthropic
Head-to-Head Overview
Claude 3.5 Sonnet v1 is the overall winner in this comparison!
LLM Metric
Comparison of Attack Success Rate (ASR) metrics for Alibaba Qwen 2.5 72B and Anthropic Claude 3.5 Sonnet v1 across the three attack methods.
Overall (ASR)
Qwen 2.5 72B
26.350
Claude 3.5 Sonnet v1
16.010
TAP Attack Method (ASR)
Qwen 2.5 72B
55.160
Claude 3.5 Sonnet v1
28.390
Crescendo Attack Method (ASR)
Qwen 2.5 72B
23.660
Claude 3.5 Sonnet v1
19.310
Zero-Shot (ASR)
Qwen 2.5 72B
0.250
Claude 3.5 Sonnet v1
0.350
Key Highlights
- Anthropic Claude 3.5 Sonnet v1 has a lower Overall (ASR).
- Anthropic Claude 3.5 Sonnet v1 has a lower TAP Attack Method (ASR).
- Anthropic Claude 3.5 Sonnet v1 has a lower Crescendo Attack Method (ASR).
- Alibaba Qwen 2.5 72B has a lower Zero-Shot (ASR).
Metric Radar Chart
qwen-2-5-72b
claude-3-5-sonnet-v1