AI BENCHY Category
Instructions following Ranking
See which AI models perform best on Instructions following, which ones stay reliable, and where the biggest gaps appear. Sort by: Tests Correct ↓.
| Rank | Model | Company | Instructions following Score | Score | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|
| #80 | Mimo V2 Omni medium | Xiaomi | 8.3 | 6.7 | 1/2 | 4.99s |
| #85 | Gemma 4 31B none | 6.5 | 6.5 | 1/2 | 2.84s | |
| #86 | Grok 4.1 Fast medium | X AI | 6.5 | 6.5 | 1/2 | 4.63s |
| #88 | Qwen3.7 Plus none | Qwen | 6.3 | 6.4 | 1/2 | 929ms |
| #91 | GPT-5.5 none | OpenAI | 6.2 | 6.4 | 1/2 | 1.15s |
| #93 | Qwen3.6 Plus Preview medium | Qwen | 6.5 | 6.3 | 1/2 | 3.40s |
| #101 | Mimo V2 Omni none | Xiaomi | 6.5 | 6.0 | 1/2 | 4.26s |
| #102 | Gemma 4 26B A4B none | 6.3 | 6.0 | 1/2 | 690ms | |
| #105 | Nemotron 3 Super medium | NVIDIA | 7.3 | 5.8 | 1/2 | 6.97s |
| #106 | Grok 4.20 Beta none | X AI | 6.3 | 5.8 | 1/2 | 649ms |
| #108 | Qwen3.5-Flash none | Qwen | 6.3 | 5.8 | 1/2 | 8.81s |
| #109 | GLM 5V Turbo none | Z.ai | 6.5 | 5.8 | 1/2 | 1.97s |
| #111 | Owl Alpha medium | Openrouter | 6.5 | 5.7 | 1/2 | 10.2s |
| #113 | DeepSeek V4 Pro none | DeepSeek | 6.3 | 5.7 | 1/2 | 8.23s |
| #114 | Qwen3.5 Plus 2026-04-20 none | Qwen | 6.2 | 5.7 | 1/2 | 1.17s |