A snappy, keyboard-centric terminal user interface for interacting with large language models. Chat with ChatGPT, Claude, Llama 3, Phi 3, Mistral, Gemma and more.

Text-To-Speech, RAG, and LLMs. All local!

Phi-3.5 for Mac: Locally-run Vision and Language Models for Apple Silicon

[ICCV 2025] Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data, and Metric Perspectives

GPU-accelerated Llama3.java inference in pure Java using TornadoVM.

Official repo for the paper PHUDGE: Phi-3 as Scalable Judge. Evaluate your LLMs with or without custom rubric, reference answer, absolute, relative and much more. It contains a list of all the available tool, methods, repo, code etc to detect hallucination, LLM evaluation, grading and much more.

EmbeddedLLM: API server for Embedded Device Deployment. Currently support CUDA/OpenVINO/IpexLLM/DirectML/CPU

這是一本 Phi-3 的入門書。Phi-3 是由微軟開發的一系列開放 AI 模型。Phi-3 模型是目前最強大且具成本效益的小型語言模型（SLM），在各種語言、推理、程式碼和數學基準測試中，表現優於同尺寸和更大尺寸的模型。

Fine-tuning Large Language Models (LLMs) for Text Classification Task

This project offers a user-friendly interface that allows users to easily create stories and enrich them with visuals. It supports creativity with story creation and visualisation features.

A distributed training framework for large language models powered by Lightning.

This repository demonstrates a machine learning pipeline for detecting MITRE ATT&CK techniques from logs and enriching the output using a local LLM.

The LLM Dataset Generator is an open source tool for generating text data compatible with various language models supported by LangChain. You can customize it to meet your specific needs, making it a valuable resource for researchers, developers, and organizations working on NLP applications.

Local RAG based LLM Pipeline for Grammar teaching.

AI Discord Bot (GEMM-X) is an intelligent assistant for Discord, leveraging AI technologies from multiple providers to generate images, create music, produce speech, and more. It supports custom personality settings and advanced user/server configurations.

Official implementation of "DZ-TDPO: Non-Destructive Temporal Alignment for Mutable State Tracking". SOTA (99.4% Win Rate) on Multi-Session Chat with negligible alignment tax.

Official site for OmniBot - Run LLMs natively & privately in your browser

Robotics navigation solution (on example of Pioneer 3-DX mobile robot in CoppeliaSim) powered by Phi-3 small language model and AMD AI PC.

⚗️ Phi-3-mini 3.8B instruct model repository