AI BENCHY Category Failures
Instructions following: No answer
Instructions following
No answer
See which AI models are most likely to hit No answer on Instructions following, so you can spot weak points faster. Sort by: Tests Correct ↑.
Failure Reasons
| Rank | Model | Company | No answer Count | Category Score | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|
| #53 | Gemini 3.1 Flash Lite high | 1 | 7.3 | 1/2 | 23.3s | |
| #161 | Qwen3.5-9B medium | Qwen | 1 | 6.5 | 1/2 | 5.75s |