Local & Self-Hosted AI Tools

Scarf gives people running the Hermes AI agent a native interface for conversations, project work and host management. It requires an existing Hermes installation; the iPhone companion, ScarfGo, connects directly to your hosts over SSH.

Super Agent Party is an AI companion platform for people who want a desktop character that can also carry out tasks. It combines character conversations with agents that operate applications and interact through messaging or live streams.

OpenSandbox gives developers isolated environments for AI agents that need to execute code or interact with browsers and developer tools. It runs on your own infrastructure, locally with Docker or on a Kubernetes cluster.

PotPlayer ChatGPT Translate translates existing subtitles inside PotPlayer during video playback. It's for viewers who want AI translation with a choice of a local model or a hosted API.

Embedding Atlas helps researchers and developers inspect embedding datasets and the records behind them. It makes clusters, outliers, and relationships visible alongside the data's metadata.

TalkToPaste is a voice dictation utility for macOS and Windows that runs speech recognition on your computer. The official site links to the VoiceToClipboard repository, which documents the local implementation and installation scripts. Transcription works offline after the speech model has been downloaded.

Slotstream runs Qwen3.8-Flash-Next on Apple Silicon Macs with too little RAM to hold the whole model. It's aimed at Macs with 16 to 64 GB of memory, for people who want local chat, answers about pictures or a backend for coding assistants.

ChuanhuChatGPT is a self-hosted web interface for people who want to chat with local LLMs and cloud models in one application. It also supports document questions and automated agent tasks.

HitPaw FotorPea is a proprietary AI photo enhancer for Windows and macOS that combines photo restoration with image generation and editing. It's aimed at photographers, online sellers and people repairing family photos who want automatic corrections without extensive retouching experience.

HunyuanDiT is a local text-to-image model for people generating images from Chinese or English prompts. It handles detailed descriptions and Chinese cultural references, including poetry, idioms, food and architecture.

Nativ is a macOS app for people who want AI chat and coding tools to use models running on their own Mac. It combines a chat workspace, model library, and local inference server.

Tolaria is a desktop knowledge base for personal notes, company documentation, and assistant memory. It keeps your material in ordinary Markdown files that other editors and tools can read.

Fortress is a self-hosted Chromium fork for developers whose scrapers and AI agents encounter browser fingerprint checks. It changes fingerprint behavior inside the browser engine to help automated sessions access protected pages.

SnowChat is a self-hosted text-to-SQL app for people who want to ask questions about their Snowflake data in plain language. It generates SQL and returns results through a Streamlit chat interface, so analysts and business users can explore data without writing every query themselves. A hosted demo is also available.

ColBERT is a self-hosted neural search library for developers building document search or retrieval for AI applications. It finds relevant passages in your own text collections and stores their search representations on disk. The Python project is open source under the MIT license.

python-audio-separator splits recordings into separate vocal and instrumental tracks on your own hardware. It's a Python package for karaoke creators and developers who want to process audio in scripts or applications.

CubeSandbox is a self-hosted sandbox service for teams that need AI agents to execute code on their own infrastructure. Each sandbox has its own operating system kernel inside a hardware-isolated MicroVM.

AgentScope is a Python framework for developers building AI agent applications on their own servers. It includes an SDK, a FastAPI backend and a web UI for applications with multiple users and separate sessions.

Hold Speak is a macOS menu bar app for people who want to dictate prompts to AI agents or enter text in other apps. It runs Whisper locally on your GPU through WhisperKit and CoreML, so local transcription works offline and keeps speech on your Mac. Push-to-talk dictation inserts the transcript into the focused text field.

mergekit combines pretrained language models into one model for developers and researchers who want to combine specialized skills or balance model behaviors. It works directly with model weights, without additional training or access to the original training data.

GPT Crawler collects website text into knowledge files for people building AI assistants around documentation or other web content. It prepares the material a custom GPT or assistant can use to answer questions about a site.

VideoLingo is a video translation app for creators and educators producing bilingual subtitles or dubbed videos. Its self-hosted browser interface brings transcription, translation and subtitle preparation into one workflow, though the selected providers determine which processing stays local.

PenguinHarness builds runnable AI agent applications from plain-language requirements on your computer or server. It's for developers who want agents to handle app creation, evaluation, tuning and deployment.

LiteRT is Google's framework for developers adding machine learning and generative AI to apps that run on users' devices. It succeeds TensorFlow Lite and supports local inference without a server round trip.

HunyuanCustom generates videos of specific people or objects for creators and researchers who need subjects to stay recognizable across frames. It builds on HunyuanVideo and supports generation on your own hardware.

Prodigy is an annotation tool for developers and research teams building training and evaluation datasets for custom AI models. It runs on your own machines and provides a web interface where annotators can label data without programming knowledge.

KubeAI is a Kubernetes operator for teams serving AI models on their own infrastructure or cloud clusters. It manages model servers and adjusts capacity to demand, including starting models from zero running replicas.

Agent Console helps developers and teams monitor Claude Code and Codex sessions through a local dashboard. It reads the agents' transcripts on your machines so you can compare usage and spot sessions that need attention.

MOSS is a conversational language model from Fudan University for developers and researchers who want to run and adapt a local LLM. It includes model weights, dialogue datasets and fine-tuning code.

Superwhisper turns spoken words into text in the app you're using. It suits people who prefer dictating messages, documents or requests to coding assistants over typing them.

BeeAI Framework is for developers building AI agents and systems where several agents work together. It lets developers enforce rules on agent behavior while retaining the model's ability to reason.

Safetensors is a file format and library for developers storing and loading AI model weights on their own hardware or servers. It addresses the arbitrary code execution risk of pickle-based PyTorch files.

Voci is a local text-to-speech reader for Windows 10 and 11 that speaks in voices cloned from short audio recordings. It's for people who want to listen to documents or pasted text in a chosen voice without sending the content to a cloud service. After the initial downloads, it works offline and doesn't require an account or API key.

Tianshou is a Python reinforcement learning library built on PyTorch and Gymnasium. It's for researchers developing learning algorithms and developers training agents in custom environments. Its two API levels let you choose between a high-level interface for applying existing algorithms and modular interfaces for changing how an algorithm learns.

Whisper is an open source speech recognition model for transcribing recordings on your own hardware. It's for developers building voice features and people working with multilingual audio.

APPFL is an MIT-licensed Python framework for researchers and data scientists who want to train machine learning models across distributed clients while keeping training data decentralized. It supports user-defined models and lets teams test privacy-preserving federated learning methods in simulations or real distributed experiments.

MCP Toolbox for Databases connects coding assistants and AI agents to databases. It's for developers who want assistants to inspect data during development or use defined database actions in applications.

TEN Framework is a self-hosted framework for developers building voice agents and multimodal conversational applications. Its default assistant example depends on external services even when the application runs on your hardware.

TorchMD-NET is an open-source Python library for researchers who train neural network potentials for molecular dynamics. It exposes its models as PyTorch modules and connects them to GPU-accelerated simulation software, including ACEMD, OpenMM and TorchMD. It's available under the MIT license.

Gradio builds browser interfaces around machine learning models, APIs and Python functions. It's for developers and researchers who want people to try an AI application without writing a separate JavaScript or CSS frontend.

Obsidian Copilot is an AI plugin for writers and researchers who want to work with the notes already in their vault. Its agents can find relevant material, draft notes and organize research inside Obsidian.

ai-memory keeps project context for developers who switch coding assistants or machines. It preserves unfinished work, failed approaches and open questions so the next session can pick up the task.

Ollama runs open models for people who want local AI applications, model chat or a backend for coding assistants. It provides a command-line interface and an API, with a choice between inference on your hardware and its cloud service.

AgentENV runs isolated environments for AI agents across your own servers. It's for teams building parallel agent workflows or reinforcement learning systems that need many separate sandboxes.

Deep Agents is an AI agent framework for developers building assistants that carry out multi-step work. It provides planning and task execution defaults that you can customize without forking the project.

Label Studio is a platform for teams preparing training datasets or reviewing AI outputs. It brings human annotation and evaluation into a browser interface that can run locally or on your own server.