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 ↑.
216/216
Filter models
No models match the current search and filters.
| Rank | Model | Company | Instructions following Score | Score | Total Cost | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|---|
| #26 | Claude Sonnet 5 medium | Anthropic | 9.9 | 8.3 | $0.922 | 2/2 | 3.10s |
| #28 | Gemini 2.5 Flash medium | 9.8 | 8.2 | $0.643 | 2/2 | 2.62s | |
| #29 | GPT-5 Mini medium | OpenAI | 10.0 | 8.1 | $0.237 | 2/2 | 11.6s |
| #32 | Inkling high | Thinkingmachines | 9.8 | 8.0 | $1.006 | 2/2 | 7.00s |
| #33 | Step 3.7 Flash medium | Stepfun | 9.8 | 8.0 | $0.515 | 2/2 | 1.83s |
| #34 | GPT-5.2 Chat none | OpenAI | 9.8 | 8.0 | $0.604 | 2/2 | 5.51s |
| #35 | GLM 5.2 high | Z.ai | 10.0 | 8.0 | $0.817 | 2/2 | 4.26s |
| #36 | Inkling medium | Thinkingmachines | 9.8 | 8.0 | $0.391 | 2/2 | 6.17s |
| #37 | Kimi K3 max | Moonshot AI | 10.0 | 8.0 | $3.112 | 2/2 | 7.66s |
| #38 | GPT-5.6 Terra high | OpenAI | 10.0 | 8.0 | $1.055 | 2/2 | 2.14s |
| #39 | Seed-2.0-Lite medium | Bytedance Seed | 10.0 | 7.9 | $0.234 | 2/2 | 7.26s |
| #40 | Qwen3.7 Plus medium | Qwen | 10.0 | 7.9 | $0.267 | 2/2 | 16.1s |
| #41 | Qwen3.6 Plus medium | Qwen | 10.0 | 7.8 | $0.405 | 2/2 | 7.54s |
| #42 | GLM 5.2 medium | Z.ai | 9.9 | 7.8 | $0.187 | 2/2 | 7.90s |
| #43 | GPT-5.6 Terra medium | OpenAI | 10.0 | 7.8 | $0.676 | 2/2 | 1.43s |