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.
| Rank | Model | Company | Instructions following Score | Score | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|
| #52 | Claude Sonnet 4.6 medium | Anthropic | 10.0 | 7.4 | 2/2 | 2.61s |
| #54 | GPT-5 Mini medium | OpenAI | 10.0 | 7.3 | 2/2 | 11.6s |
| #58 | Gemini 3.1 Flash Lite Preview none | 10.0 | 7.2 | 2/2 | 1.13s | |
| #60 | Kimi K2.6 medium | Moonshot AI | 10.0 | 7.2 | 2/2 | 12.5s |
| #61 | Gemini 3.1 Flash Lite low | 10.0 | 7.2 | 2/2 | 1.52s | |
| #64 | MiMo-V2-Flash medium | Xiaomi | 10.0 | 7.2 | 2/2 | 4.28s |
| #66 | Qwen3.5-35B-A3B medium | Qwen | 10.0 | 7.1 | 2/2 | 24.4s |
| #69 | Claude Opus 4.6 medium | Anthropic | 10.0 | 7.0 | 2/2 | 2.43s |
| #72 | DeepSeek V3.2 medium | DeepSeek | 10.0 | 7.0 | 2/2 | 35.8s |
| #73 | Seed-2.0-Mini medium | Bytedance Seed | 10.0 | 6.9 | 2/2 | 17.5s |
| #76 | Kimi K2.5 medium | Moonshot AI | 10.0 | 6.8 | 2/2 | 92.5s |
| #78 | Qwen3.6 27B medium | Qwen | 10.0 | 6.8 | 2/2 | 38.0s |
| #81 | Mercury 2 medium | Inception | 10.0 | 6.6 | 2/2 | 1.07s |
| #82 | Hy3 preview high | Tencent | 10.0 | 6.6 | 2/2 | 34.4s |
| #83 | Step 3.5 Flash none | Stepfun | 10.0 | 6.6 | 1/1 | 9.30s |