Puzzle Solving: Did not follow instructions
Puzzle Solving
Did not follow instructions
See which AI models are most likely to hit Did not follow instructions on Puzzle Solving, so you can spot weak points faster. Sort by: Total Cost ↓.
Failure Reasons
157/157
Filter models
No models match the current search and filters.
| Rank | Model | Company | Did not follow instructions Count | Category Score | Total Cost | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|---|
| #136 | Muse Glimmer 30B high | Meta | 1 | 0.6 | $0.506 | 1/3 | 64.8s |
| #316 | Kimi K2.6 none | Moonshot AI | 1 | 0.3 | $0.486 | 0/3 | 1.40s |
| #99 | Gemini 4 Argon low | 1 | 0.9 | $0.438 | 2/3 | 2.18s | |
| #137 | Nemotron 3.5 Super VL medium | NVIDIA | 1 | 0.8 | $0.437 | 2/3 | 11.8s |
| #157 | Inkling Small high | Thinkingmachines | 1 | 0.5 | $0.398 | 0/3 | 14.4s |
| #36 | DeepSeek V4.1 Flash 1017 low | DeepSeek | 1 | 0.7 | $0.355 | 1/3 | 2.54s |
| #341 | Mimo V2 PRO medium | Xiaomi | 1 | 0.6 | $0.333 | 1/3 | 5.08s |
| #248 | MiMo-V2.5-Pro medium | Xiaomi | 1 | 0.7 | $0.332 | 1/3 | 5.31s |
| #351 | MiniMax M2.7 medium | Minimax | 1 | 0.6 | $0.326 | 1/3 | 24.9s |
| #269 | GLM 5 Turbo medium | Z.ai | 1 | 0.9 | $0.323 | 2/3 | 5.23s |
| #84 | GPT-5 Mini medium | OpenAI | 1 | 0.6 | $0.315 | 1/3 | 15.2s |
| #238 | Qwen3.5-122B-A10B none | Qwen | 1 | 0.4 | $0.308 | 0/3 | 1.00s |
| #153 | DeepSeek V4 Flash 0731 low | DeepSeek | 1 | 0.8 | $0.306 | 2/3 | 9.08s |
| #203 | Inkling low | Thinkingmachines | 1 | 0.6 | $0.293 | 1/3 | 2.97s |
| #304 | Mistral Large 4 none | Mistral | 1 | 0.8 | $0.290 | 2/3 | 19.3s |