Sample to envision intelligent apps with Microsoft's Copilot stack for AI-infused product experiences.

Phi2-Chinese-0.2B 从0开始训练自己的Phi2中文小模型，支持接入langchain加载本地知识库做检索增强生成RAG。Training your own Phi2 small chat model from scratch.

Inferflow is an efficient and highly configurable inference engine for large language models (LLMs).

Examples of RAG using Llamaindex with local LLMs - Gemma, Mixtral 8x7B, Llama 2, Mistral 7B, Orca 2, Phi-2, Neural 7B

LLM inference in Fortran

Examples of RAG using LangChain with local LLMs - Mixtral 8x7B, Llama 2, Mistral 7B, Orca 2, Phi-2, Neural 7B

Collection of Basic Prompt Templates for Various Chat LLMs (Chat LLM 的基础提示模板集合)

Test server code for Phi-2 model. support OpenAI API spec

Microsoft Phi 2 Streamlit App, deployed on HuggingFace Spaces is based on the Microsoft Phi 2 small language model (SLM) for text generation.

Build a Conversational AI System that can answer questions by retrieving the answers from a document.

Co:Here Inference configurations

Examples of RAG using Llamaindex with local LLMs in Linux - Gemma, Mixtral 8x7B, Llama 2, Mistral 7B, Orca 2, Phi-2, Neural 7B

Simple LLM Rest API using Rust, Warp and Candle. Dedicated for quantized version of either phi-2 ( default) , Mistral, or Llama. Work using CPU or CUDA

Fine tune Phi 2 for persona grounded chat

Code submission for the "Specializing Large Language Models for Telecom Networks by ITU AI/ML in 5G" challenge

This repository contains a Python script for a Telegram bot that integrates with OpenAI's API or other compatible REST APIs (such as Jan https://jan.ai/). It's designed to provide an interactive AI experience through Telegram, using simple chat functionalities.

This repository contains the source code used for finetuning the LLM phi-2 with several frameworks, such as DPO.

FineTune Microsoft Phi-2 with your own data

This project is an AI-powered algebra tutor using the Phi-3 Mini model. It provides personalized learning through interactive chat, adapting to the student's level and offering detailed step-by-step solutions. Built with Streamlit for an engaging educational experience.