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

Narrative Understanding on HS

87.8First-Token Accuracy

Phi-4-14B

Updated 1mo ago

Evaluation Results

Method	Links
Phi-4-14B 2025.05		87.8
Ministral-8B 2025.05		86.5
Phi-4-14B 2025.05		85.6
Ministral-8B 2025.05		81.1
Mistral-Nemo-12B 2025.05		76.4
Gemma-2-9B 2025.05		69.8
Llama-3.1-8B 2025.05		69
Ministral-8B 2025.05		68.6
Qwen-2-7B 2025.05		65.2
Qwen-2-7B 2025.05		65
Gemma-2-9B 2025.05		63.6
Phi-4-14B 2025.05		61.2
Gemma-7B 2025.05		56.3
Qwen-2-7B 2025.05		53.4
Mistral-Nemo-12B 2025.05		52.6
Llama-3.1-8B 2025.05		52.2
Mistral-Nemo-12B 2025.05		51.6
Llama-3.1-8B 2025.05		49.2
Gemma-7B 2025.05		43.9
Gemma-2-9B 2025.05		43
Zephyr-7B 2025.05		40
Gemma-7B 2025.05		36.3
Zephyr-7B 2025.05		36.1
Zephyr-7B 2025.05		35