Instructions following Ranking
See which AI models perform best on Instructions following, which ones stay reliable, and where the biggest gaps appear. Sort by: Total Cost ↓.
461/461
Filter models
No models match the current search and filters.
| Rank | Model | Company | Instructions following Score | Score | Total Cost | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|---|
| #179 | Qwen3.8 27B medium | Qwen | 1.0 | 0.7 | ~$0.073 | 2/2 | 2.54s |
| #149 | Qwen3.7 Flash low | Qwen | 0.8 | 0.8 | $0.071 | 1/2 | 9.88s |
| #381 | Qwen3 Coder Next medium | Qwen | 0.6 | 0.5 | $0.071 | 1/2 | 7.49s |
| #396 | KAT Coder AIR V2.5 default | Kwaipilot | 1.0 | 0.4 | $0.070 | 2/2 | 1.75s |
| #299 | Gemini 3.1 Flash Lite none | 1.0 | 0.6 | $0.069 | 2/2 | 859ms | |
| #411 | Grok 4.1 Fast medium | X AI | 0.7 | 0.4 | $0.069 | 1/2 | 4.63s |
| #383 | GPT-5.4 Nano none | OpenAI | 0.6 | 0.5 | $0.068 | 1/2 | 784ms |
| #216 | Solar Pro 4 xhigh | Upstage | 0.8 | 0.7 | $0.067 | 1/2 | 51.3s |
| #366 | Qwen3 Coder Next default | Qwen | 0.6 | 0.5 | $0.067 | 1/2 | 7.78s |
| #160 | GPT-5.6 Luna medium | OpenAI | 1.0 | 0.7 | $0.065 | 2/2 | 2.38s |
| #426 | Command A+ medium | Cohere | 0.3 | 0.3 | $0.064 | 0/2 | 103.0s |
| #147 | Claude Haiku 5.5 medium | Anthropic | 1.0 | 0.8 | $0.063 | 2/2 | 3.01s |
| #182 | Qwen3.7 Flash high | Qwen | 1.0 | 0.7 | $0.063 | 2/2 | 9.41s |
| #83 | Claude Haiku 5.5 low | Anthropic | 1.0 | 0.9 | $0.061 | 2/2 | 2.20s |
| #336 | MiMo-V2.6-Flash none | Xiaomi | 0.7 | 0.5 | $0.060 | 1/2 | 810ms |