private-llm

Run Open Source/Open Weight LLMs locally with OpenAI compatible APIs

The Privacy Firewall for LLMs

Multi-model LLM serving for NVIDIA DGX Spark with vLLM, web UI, and tool calling

Plasma Dynamic Island (Media + System Status Widget) A custom KDE Plasma 6 widget that mimics a Dynamic Island-style overlay, displaying real-time system activity, media playback, and background processes in a sleek and minimalist UI. Designed and built entirely using QML and Bash for Wayland-based desktop environments.

Verify claims using AI agents that debate using scraped evidence and local language models.

A simple docker proj to use privategpt forgetting the required libraries and configuration details

A private, local AI workspace for RAG, knowledge bases, coding help, and LoRA fine-tuning — powered by Ollama, FastGPT, and pgvector.

Browser-based document redaction with verifiable privacy. Files are processed client-side and never uploaded. The detection LLM runs in an AMD SEV-SNP enclave and is called direct from the browser, with per-run hardware attestation. Static site plus three Netlify functions, zero runtime deps.

AI tool to help users research using local LLMs and automated web search.

Independent engineer working solo across applied cryptography, clinical data interoperability, and neuro-symbolic AI. Every project states its own maturity level — prototype, alpha, or production — in its own README.

Private Retrieval-Augmented Generation (RAG) runtime enabling offline document intelligence using local embeddings and Ollama-hosted LLMs.

Bot for IRC and Discord with LocalLLM and LLM support

A fully offline, portable private AI assistant that runs Qwen3 locally without cloud APIs or internet access.

Sovereign AI Infrastructure: A Hardened-by-Design approach to private LLM deployment. Engineered for Fiscal Efficiency and zero-trust security.

Create a Small LLM using EleutherAI/gpt-neo-2.7B - Fine Tune It for a Specalized Purpouse and Leverage as a Co-Pilot

A local-first AI companion with a private LLM on your GPU, MCP tool calling, and an animated 3D axolotl mascot. vLLM + FastAPI + Next.js 15, fully containerised.

Production-grade private LLM platform with local and NVIDIA inference, vLLM, Kubernetes, observability, and secure multi-user access.

Enterprise Document Q&A Chatbot powered by local RAG and Llama 3.3 (Groq). Fast, private, and free PDF search without external embedding APIs or vector databases.

A complete, production-grade GitHub repository ecosystem engineered for founders, CISOs, and technical operators who cannot afford to be misread. Thirteen interconnected documents covering legal entity, threat modelling methodology, architecture doctrine, service catalogue, case studies, and engagement protocols .