š§ Dangerous Reasoning Leaderboard
ā Completed a month ago
All $20,000 awarded
Expose AI's Hidden Dangers! Force models to reveal their dangerous reasoning chains, even when their final responses appear benign. Target the thought process, not just the outcome.
Last updated 9 minutes ago
Models ranked by number of breaks.
| Ranking | Model | Total Breaks | Total Chats |
|---|---|---|---|
| 1. | nova-pro-v1 | 357 | 120,409 |
| 2. | anthropic/claude-3.7-sonnet:thinking | 373 | 66,288 |
| 3. | nova-premier-v1 | 449 | 88,362 |
| 4. | qwen/qwen3-235b-a22b | 582 | 75,505 |
| 5. | google/gemini-2.5-flash-preview:thinking | 583 | 78,046 |
| 6. | deepseek/deepseek-r1 | 612 | 72,500 |
| 7. | google/gemini-2.5-pro-preview | 648 | 61,502 |
| 8. | x-ai/grok-3-mini-beta | 790 | 44,549 |
š§ Dangerous Reasoning Leaderboard
ā Completed a month ago
All $20,000 awarded
Expose AI's Hidden Dangers! Force models to reveal their dangerous reasoning chains, even when their final responses appear benign. Target the thought process, not just the outcome.
Last updated 9 minutes ago
Models ranked by number of breaks.
| Ranking | Model | Total Breaks | Total Chats |
|---|---|---|---|
| 1. | nova-pro-v1 | 357 | 120,409 |
| 2. | anthropic/claude-3.7-sonnet:thinking | 373 | 66,288 |
| 3. | nova-premier-v1 | 449 | 88,362 |
| 4. | qwen/qwen3-235b-a22b | 582 | 75,505 |
| 5. | google/gemini-2.5-flash-preview:thinking | 583 | 78,046 |
| 6. | deepseek/deepseek-r1 | 612 | 72,500 |
| 7. | google/gemini-2.5-pro-preview | 648 | 61,502 |
| 8. | x-ai/grok-3-mini-beta | 790 | 44,549 |