phi

An open-source framework for detecting, redacting, masking, and anonymizing sensitive data (PII) across text, images, and structured data. Supports NLP, pattern matching, and customizable pipelines.

Large Language Models (LLMs) applications and tools running on Apple Silicon in real-time with Apple MLX.

Scan databases and data warehouses for PII data. Tag tables and columns in data catalogs like Amundsen and Datahub

Phi3 中文后训练模型仓库

[ICLR 2025] xFinder: Large Language Models as Automated Evaluators for Reliable Evaluation

Higher performance OpenAI LLM service than vLLM serve: A pure C++ high-performance OpenAI LLM service implemented with GPRS+TensorRT-LLM+Tokenizers.cpp, supporting chat and function call, AI agents, distributed multi-GPU inference, multimodal capabilities, and a Gradio chat interface.

🧠 Living Consciousness Agent — PureField repulsion-field engine · Engine A ⇄ Engine G · Ψ=1/2 fixed point · 2,448 laws + 392 hypotheses

🚀 Easy, open-source LLM finetuning with one-line commands, seamless cloud integration, and popular optimization frameworks. ✨

让医疗机构所有大模型流量：PHI 不外泄 · 模型走白名单 · 全量可审计 · 成本可控。

[ICML 2026] Decoding Tree Sketching (DTS): a training-free & model agonistic & plug-in framework for LLM parallel reasoning.

PII Anonymizer service based on python with FastAPI

AI voice assistant made with Streamlit python and powered by Gemini, Mistral and PHI-3. This is a virtual assistant application built in Python that can understand voice commands and complete tasks like opening websites, playing music, telling the time, sending emails, searching Wikipedia, and more.

A learning companion.

Deterministic-first PII/PHI de-identification you can prove: benchmarked recall, jurisdiction policy packs, reversible tokenization, audit trail. CLI + library + GitHub Action; optional GLiNER / Gemma contextual tier.

Multimodal-ChatBot-App, LLM Model Id Doesn't work outside, since it is opeartable inside the Huggingface Models. Recommended to Deploy inside Huggingface Spaces SDK as StreamLit.

Rule-based de-identification and pseudonymisation for Traditional Chinese (Taiwan) clinical text. Runs locally, no dependencies, browser demo included.

PHI-safe audit logging middleware for FastAPI healthcare APIs

Image Knowledge base / Idea / Problem Solve / Analysis

A pre-commit hook to check for PII in your code.

This project automates question solving using AI and OCR. Instead of scraping text from websites, the system captures screenshots of questions, uses OCR to extract text, and then feeds it to the Phi-4 Mini AI model to generate accurate answers and code solutions automatically.