Description:

The Phi-3-Medium-4K-Instruct is a 14B parameters, lightweight, state-of-the-art open model trained with the Phi-3 datasets that includes both synthetic data and the filtered publicly available websites data with a focus on high-quality and reasoning dense properties. The model supports 4K context length (in tokens).

The model has underwent a post-training process that incorporates both supervised fine-tuning and direct preference optimization for the instruction following and safety measures. When assessed against benchmarks testing common sense, language understanding, math, code, long context and logical reasoning, Phi-3-Medium-4K-Instruct showcased a robust and state-of-the-art performance among models with less than 13 billion parameters.

The model is licensed under the MIT license

Terms of use:

By accessing this model, you are agreeing to the Terms and Conditions of the MIT License.

References(s):

• Phi-3-medium-4k-instruct Model Card
• Phi-3 blogpost

Model Architecture:

Architecture Type: Transformer

Input:

Input Format: Text

Input Parameters: None

Output:

Output Format: Text

Output Parameters: None

Software Integration:

Supported Hardware Platform(s): RTX 4090, Ada GPUs

Supported Operating System(s): Windows

Inference:

TRT-LLM Inference Engine
Windows Setup with TRT-LLM

Test Hardware:

RTX 4090