Share your thoughts, 1 month free Claude Pro on usSee more

Commonsense Reasoning on CQA

79.8First-Token Accuracy

Phi-4-14B

Updated 12d ago

Evaluation Results

Method	Links
Phi-4-14B 2025.05		79.8
Phi-4-14B 2025.05		79.6
Qwen-2-7B 2025.05		79.4
Qwen-2-7B 2025.05		79.4
Qwen-2-7B 2025.05		79.4
Gemma-2-9B 2025.05		79
Gemma-2-9B 2025.05		78.7
Llama-3.1-8B 2025.05		77.2
Llama-3.1-8B 2025.05		76.5
Mistral-Nemo-12B 2025.05		75.4
Ministral-8B 2025.05		75
Ministral-8B 2025.05		74.3
Mistral-Nemo-12B 2025.05		74.1
Llama-3.1-8B 2025.05		72.7
Phi-4-14B 2025.05		72.5
Ministral-8B 2025.05		69.6
Zephyr-7B 2025.05		69
Mistral-Nemo-12B 2025.05		68.8
Gemma-7B 2025.05		68.6
Zephyr-7B 2025.05		66
Gemma-7B 2025.05		64.7
Zephyr-7B 2025.05		54.4
Gemma-7B 2025.05		51.6
Gemma-2-9B 2025.05		39.6