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

Fine tune Phi 2 for persona grounded chat

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.

An LLM-based car problem diagnosis system that uses fine-tuned language models to analyze natural language vehicle complaints and generate structured troubleshooting guidance with safety-aware recommendations.

A modular LoRA fine-tuning pipeline for Hugging Face models with built-in data cleaning, smart early stopping, automatic mode detection, and cross-platform support.

A comprehensive toolkit for fine-tuning Microsoft's Phi-2 and Phi-3.5 language models, featuring memory-efficient training, interactive chat, and model comparison capabilities.

Edge LLM for In-Vehicle Deployment - Optimized small language model for automotive conversational AI

A novel 3D attention mechanism inspired by HDD platter geometry, featuring circular context wrapping and depth stacking to mitigate boundary effects in long-context language modeling.

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.

⚡ Enterprise-ready federated learning for LLMs. On-device personalization, privacy-preserving aggregation, LoRA adapters, real-time dashboard. Perfect for privacy-sensitive AI applications. 🌟🏆

Flask API for generating text with the Phi-2 model from Hugging Face Transformers.

Fine-tuned Phi-2 (2.7B) for NIFTY 50 options signal generation — LoRA, RAG, walk-forward evaluation, W&B experiment tracking

Medical Image Captioning with ViT-Base + Phi-2 + LoRA on ROCOv2 radiology dataset

A multi-agent LLM research system that combines planning, web search, memory, and self-critique to generate refined research reports using Phi-2.

A conversational AI chatbot for the financial technology domain using Llama with llama.cpp for local inference

Local REST API serving Phi-2 behind FastAPI, with an optional MLX LoRA adapter. Native MLX on Apple Silicon or portable PyTorch CPU in Docker; synchronous and streaming generation.

Local RAG system using ChromaDB, sentence-transformers, and Phi-2 on Apple Silicon

Demonstrates the non-deterministic behavior of LLMs by sending the same prompt multiple times and comparing the generated responses. Implemented using Python and the Hugging Face Transformers library with the Phi-2 model.