Phi Models

This page provides detailed technical specifications for the Phi model family supported by NeMo Customizer. For information about supported features and capabilities, refer to Tested Models.

Microsoft Phi-4

Property	Value
Creator	Microsoft
Architecture	Decoder-only Transformer
Description	Phi-4 is Microsoft’s most advanced small language model, designed to deliver strong reasoning capabilities while being efficient to deploy.
Max I/O Tokens	16K
Parameters	14 billion
Training Data	High-quality data with emphasis on reasoning and code
Default Name	microsoft/phi-4
HuggingFace	microsoft/phi-4

Training Options

LoRA: 2x 80GB GPU, tensor parallel size 2, micro batch size 2
Full SFT: 4x 80GB GPU, tensor parallel size 2

Deployment Configuration

LoRA:
NIM Image: nvcr.io/nim/nvidia/llm-nim:1.15.5
Full SFT:
NIM Image: nvcr.io/nim/nvidia/llm-nim:1.15.5
Additional Environment Variables:
NIM_MODEL_PROFILE: vllm