Model Overview

Terms of use:

GOVERNING TERMS: By accessing this engine, you are agreeing to NVIDIA TensorRT Engine License.

ATTRIBUTION: MIT License (https://huggingface.co/microsoft/phi-2/resolve/main/LICENSE)

Third-Party Community Consideration

This model is not owned or developed by NVIDIA. This model has been developed and built to a third-party’s requirements for this application and use case.

Description:

Phi-2 is a 2.7 billion parameter language model developed by Microsoft Research. The phi-2 model is best suited for prompts using the Question-Answer (QA) format, the chat format, and the code format. TensorRT engines optimizing Phi-2 are generated by TensorRT-Cloud.

These TensorRT engines are ready for commercial use.

References(s):

• Phi-2 on HuggingFace
  
• Microsoft Research blog post about Phi-2

Model Architecture:

Architecture Type: Transformer
Network Architecture: Phi

Input:

Input Type(s): Text
Input Format: Text
Input Parameters: Temperature, TopP
Other Properties Related to Input: None

Output:

Output Type(s): Text
Output Format: Text
Output Parameters: Max output tokens
Other Properties Related to Output: None

Inference:

TRT-LLM Inference Engine

Ethical Considerations:

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse. Please report security vulnerabilities or NVIDIA AI Concerns here.