AI BENCHY Category Failures
Coding: Did not follow instructions
Coding
Did not follow instructions
See which AI models are most likely to hit Did not follow instructions on Coding, so you can spot weak points faster. Sort by: Response Time (avg) ↑.
Failure Reasons
| Rank | Model | Company | Did not follow instructions Count | Category Score | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|
| #153 | Granite 4.1 8B none | IBM Granite | 1 | 5.2 | 0/2 | 706ms |
| #115 | MiMo-V2.5-Pro none | Xiaomi | 1 | 5.0 | 0/2 | 1.80s |
| #149 | MiMo-V2-Flash none | Xiaomi | 1 | 4.9 | 0/2 | 2.04s |
| #101 | Qwen3.5 Plus 2026-04-20 none | Qwen | 1 | 4.4 | 0/2 | 2.08s |
| #24 | Gemini 3.5 Flash minimal | 1 | 7.0 | 1/2 | 3.39s | |
| #6 | Gemini 3.5 Flash medium | 1 | 6.8 | 1/2 | 9.91s | |
| #100 | Owl Alpha medium | Openrouter | 1 | 6.6 | 1/2 | 19.1s |
| #114 | DeepSeek V3.2 none | DeepSeek | 1 | 3.1 | 0/2 | 20.9s |
| #87 | Grok 4.1 Fast medium | X AI | 1 | 2.3 | 0/1 | 23.6s |
| #63 | Claude Opus 4.6 medium | Anthropic | 1 | 7.2 | 1/2 | 29.4s |
| #74 | Laguna M.1 medium | Poolside | 1 | 4.3 | 0/1 | 35.6s |
| #80 | DeepSeek V4 Pro high | DeepSeek | 1 | 2.8 | 0/2 | 51.8s |
| #96 | Nemotron 3 Super medium | NVIDIA | 1 | 3.1 | 0/2 | 62.4s |
| #105 | Cobuddy medium | Baidu | 1 | 4.1 | 0/2 | 79.2s |
| #110 | Kimi K2.6 none | Moonshot AI | 1 | 6.8 | 1/2 | 122.8s |