AI BENCHY
Advertise here

AI BENCHY Category Failures

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: Tests Correct ↓.

Models Shown

15

Total Failures

78

Rank Model Company Did not follow instructions Count Category Score Tests Correct Response Time (avg)
#126 gpt-oss-120b none OpenAI 1 6.0 1/3 8.21s
#130 MiniMax M2.7 medium Minimax 1 5.9 1/3 24.9s
#134 GLM 5 Turbo none Z.ai 1 5.5 1/3 2.65s
#136 Elephant Alpha medium Openrouter 1 5.3 1/3 868ms
#141 Nemotron 3 Super none NVIDIA 1 5.5 1/3 2.36s
#143 MiMo-V2.5 none Xiaomi 1 5.4 1/3 2.13s
#144 GPT-5.4 Mini none OpenAI 1 5.4 1/3 836ms
#148 GPT-5.4 Nano none OpenAI 1 5.4 1/3 1.25s
#70 GPT-5.4 Nano medium OpenAI 1 4.1 0/3 3.79s
#105 Nemotron 3 Super medium NVIDIA 1 3.0 0/3 3.15s
#117 Qwen3.5-35B-A3B none Qwen 1 3.7 0/3 1.35s
#119 Cobuddy medium Baidu 1 3.6 0/3 12.8s
#124 Kimi K2.6 none Moonshot AI 1 3.1 0/3 1.40s
#128 Qwen3.6 Flash none Qwen 1 3.5 0/3 1.21s
#131 Qwen3.5-122B-A10B none Qwen 1 3.8 0/3 1.00s

Top Models by Did not follow instructions Count

Did not follow instructions Count vs Score

Top Models by Response Time (avg)

Top Models by Estimated Wasted Cost