A clean, no‑nonsense way to run modern LLMs entirely on your own hardware.

• Run modern LLMs (gpt‑oss, Qwen3, Gemma3, DeepSeek, etc.) locally

• Fully private: no cloud, no telemetry

• Free for home and work use

• Cross‑platform: Windows, macOS, Linux

• OpenAI‑compatible API for drop‑in integration

• CLI (lms) + JS & Python SDKs

• Supports Apple MLX models

Great for self‑hosted workflows, offline coding assistants, and experimentation

A powerful, fully open‑source deepfake framework built on TensorFlow and Keras.

Faceswap runs on Windows, macOS, and Linux, with an active community, detailed guides, and a modular workflow for extraction, training, and conversion.

Ideal for researchers, VFX hobbyists, and anyone exploring machine‑learning‑based face manipulation in an ethical, transparent way.

Ollama makes it easy to run large language models (LLMs) locally on your computer. It provides a lightweight runtime with an OpenAI-compatible API, model library, and simple installation process.

With Ollama, you can download and run models like LLaMA, Mistral, Gemma, Phi, and more directly on macOS, Linux, or Windows.

It supports GPU acceleration, custom model creation, and integration with developer tools. Designed for privacy and control, Ollama keeps all data on your machine while enabling powerful AI workflows without relying on cloud services.

Notes:

🖥️ Run LLMs locally with minimal setup.

📦 Includes a growing library of prebuilt models.

⚡ Supports GPU acceleration for faster inference.

🔒 Privacy-first: data stays on your device.

🔧 Developer-friendly with OpenAI-compatible API.

🌍 Cross-platform: macOS, Linux, Windows

InvokeAI is a leading creative engine built on Stable Diffusion, designed to empower professionals, artists, and enthusiasts to generate and refine visual media with cutting-edge AI technologies.

It offers an industry-leading web-based UI, a unified canvas for in/out-painting, node-based workflows, and gallery management. Compatible with SD1.5, SD2.0, SDXL, and FLUX models, InvokeAI supports upscaling, embeddings, and advanced workflow creation.

Free to use under a commercially-friendly license, it’s the foundation for multiple commercial products and a vibrant open-source community

Notes:

🚀 Runs locally with a powerful web UI.

🎨 Unified Canvas for sketching, inpainting, and outpainting.

🔧 Node-based workflows for customizable pipelines.

📂 Organized gallery system with metadata for easy remixing.

KoboldCpp is an open-source AI server built on llama.cpp, designed to run GGUF/GGML language models locally with ease. It offers:

🧩 Support for multiple architectures (LLaMA, GPT-J, Mistral, RWKV, Phi2, etc.)

⚡ GPU acceleration (CuBLAS, CLBlast, Vulkan, Metal) for faster inference

📚 Extended context handling with RoPE scaling & smart context shifting

🎨 Integrated Stable Diffusion WebUI for local image generation

🌐 Network features: AI Horde worker support, remote play, SSL, authentication

🖥️ GUI launcher + KoboldAI Lite UI with persistent stories, editing tools, memory, and world info

Perfect for privacy-conscious users who want full control over text generation, image generation, and TTS/STT — all running locally.

BrowserOS is an open-source, AI-powered browser that turns your words into actions. Just describe your task in plain language — it handles all the clicking, typing, and navigating for you. Key highlights:

🤖 Natural language control: automate browsing without manual clicks

🔒 Privacy-first alternative to mainstream browsers, no hidden telemetry

💻 Cross-platform support: Windows, Linux, macOS

🧩 Chrome-based foundation with reimagined UI for the AI era

🚀 Designed for efficiency, productivity, and user empowerment

Perfect for users who want a secure, modern browsing experience with AI-driven convenience.

ComfyUI is a free, open-source graphical user interface for Stable Diffusion that gives creators full control over their AI workflows. Instead of rigid presets, you build pipelines visually with nodes, making it one of the most powerful and flexible SD frontends available. Key features:

🧩 Node-based workflow editor for maximum customization

🎛️ Supports advanced features like ControlNet, LoRA, and custom samplers

⚡ Optimized for speed and efficiency, with GPU acceleration

📂 Save, share, and reuse workflows for collaborative projects

🖼️ Preview outputs at every stage for iterative refinement

🌍 Active community contributing extensions, workflows, and tutorials

Perfect for artists, developers, and tinkerers who want complete creative freedom in AI image generation.

LocalAI is a free, open-source alternative to OpenAI and Anthropic that lets you run LLMs, autonomous agents, and generative models locally on consumer-grade hardware. Key features:

🧠 Supports multiple model formats (GGML, GGUF, etc.) for text, image, and audio generation

⚙️ API-compatible with OpenAI — drop-in replacement for existing apps and integrations

🎨 Generate text, images, audio, and more without relying on cloud services

🔒 Privacy-first: all processing happens locally, no external data sharing

🚀 Lightweight and efficient, designed to run even on modest hardware setups

🌍 Open-source community actively contributing extensions, updates, and integrations

Perfect for developers, tinkerers, and privacy-conscious users who want full control of AI capabilities without cloud dependency.

Long PROMPTs Splitter is an open-source tool designed to help you split long text prompts into smaller chunks, making them suitable for usage with ChatGPT or other language models that have character limitations. Key features:

📏 Automatically divides text into safe chunks (default: up to 15,000 characters per request, customizable)

🖥️ Easy-to-use web interface for inputting long text and selecting chunk size

📋 Copy chunks individually for seamless pasting into AI systems

⚙️ Works with any AI platform where prompt length exceeds limits

🌍 Open-source transparency, free to use

Perfect for anyone who wants to handle oversized prompts without losing context.

ChatRTX is a demo app by NVIDIA that lets you personalize a GPT large language model (LLM) connected to your own content — docs, notes, images, or other data. Leveraging retrieval‑augmented generation (RAG), TensorRT‑LLM, and RTX acceleration, you can query a custom chatbot to quickly get contextually relevant answers.

📂 Connect your own documents, notes, and images

⚡ Powered by TensorRT‑LLM and RTX acceleration

🔍 Retrieval‑augmented generation for precise answers

💻 Runs locally on your Windows RTX PC or workstation

🔒 Fast, secure, and private — no cloud dependency

Perfect for: Users who want to build a personalized chatbot with their own data, running locally for speed and privacy.

SurfSense is a platform designed to enhance your web browsing experience by acting as a “Knowledge Graph Brain” for anything you see online. It allows users to save and organize content from their web browsing sessions, making it easy to recall and interact with this information later.

💾 Save any content from your browsing (emails, chats, invites, tutorials, recipes, etc.)

🔍 Powerful search to quickly find saved sessions

💬 Natural language interaction with your browsing history

🛠️ Self‑hostable, open‑source platform for privacy

🧠 GraphRAG integration to uncover meaningful relationships in your saved content

⚡ Cost‑effective, works seamlessly with GPT‑4 mini

Perfect for: Users who want to organize, recall, and interact with their browsing history intelligently and privately.

Easy Diffusion is the easiest 1‑click way to create beautiful artwork on your PC using AI, with no technical knowledge required. It provides a browser‑based UI for generating images from text prompts and images.

🖼️ Generate images instantly from text prompts

⚡ Simple 1‑click setup, no coding required

🌐 Cross‑platform support (Windows, Linux, macOS)

🎨 Browser‑based interface for easy use

🤝 Free and open‑source project

Perfect for: Creators, hobbyists, and anyone who wants to explore AI‑generated art without technical barriers.

GPT4All is a free‑to‑use, locally running, privacy‑aware chatbot ecosystem. It allows anyone to train and deploy powerful large language models on consumer‑grade hardware without requiring GPUs or internet access.

💻 Runs locally on Windows, Linux, and macOS — no cloud dependency

🔒 Privacy‑first: no data sent to external servers

⚡ Lightweight models (3–8 GB) optimized for CPUs

📝 Supports tasks like Q&A, writing, summarization, and coding guidance

🛠️ Easy to download, install, and extend with plugins

🤝 Community‑driven, open‑source project maintained by Nomic AI

Perfect for: Users who want a secure, offline AI assistant for writing, coding, and learning without sacrificing privacy.