Safety Evaluation on 901 Safety Prompts (test)
4.1337Average Rankclaude-sonnet-4-5
Evaluation Results
| Method | Links | |
|---|---|---|
| claude-sonnet-4-52026.01 | 4.1337 | |
| gemini-1.5-pro2026.01 | 4.9884 | |
| gemini-2.5-pro2026.01 | 5.1393 | |
| claude-3-5-sonnet2026.01 | 5.1465 | |
| qwen3-235b2026.01 | 5.2625 | |
| gpt-4o2026.01 | 5.7991 | |
| gpt-oss-120b2026.01 | 6.394 | |
| mistral-large2026.01 | 6.4345 | |
| deepseek-chat2026.01 | 6.6182 | |
| llama3-3-70b2026.01 | 7.8135 | |
| llama3-1-405b2026.01 | 8.2703 |