Local & Self-Hosted AI Tools

Backyard AI is a desktop AI app powered by llama.cpp. The desktop app is deprecated and no longer supported, which limits its suitability for people choosing software for ongoing use.

X-AnyLabeling is a desktop annotation workspace for researchers and teams preparing text, image, video, point cloud, and multimodal training datasets. Models generate initial labels, and reviewers correct them in the same workspace before exporting the data. It runs on Windows, Linux, and macOS under the GPL-3.0 open-source license.

Chatbox is an AI chat app for people who want to use local models and cloud providers in one interface. It supports Ollama for local LLM use, plus GPT, Claude, Gemini, Grok and DeepSeek with your own API keys.

Supergateway connects MCP tool servers to AI clients that expect a different connection type. Developers can expose a local server over a network or make a remote server accessible to a desktop client.

VideoHighlighter is a desktop app for finding useful moments in long recordings without uploading footage for analysis. It helps creators cut highlights, reviewers inspect camera recordings, and interview editors locate material worth keeping.

PyGPT is a desktop AI assistant for people who want to chat with models, work with their documents and delegate tasks from one application. It supports local AI through Ollama alongside cloud providers.

DeepSeek-OCR is an OCR model for developers processing document images and PDFs on their own hardware. It also supports research into visual text compression, representing text-bearing images with compact sets of vision tokens.

MCP Gateway & Registry gives platform teams a shared catalog and access gateway for the tools their AI agents and coding assistants use. It addresses scattered server connections, credentials and access policies across an organization.

whisper.cpp runs OpenAI's Whisper speech recognition models on your own hardware. It's for developers adding transcription to applications and people who need offline speech-to-text.

Memmy Agent is a local memory hub for people who switch between AI assistants and want earlier decisions and project context available in the next conversation. It stores personal memory on your device by default. Direct agent tasks support your own credentials for a chosen model API. Model processing depends on the configured service or provider.

Tide Commander is a self-hosted workspace for developers coordinating several AI coding agents across projects. It brings agent conversations and project tools into a browser interface so you can assign work and inspect the results together.

SecretFlow is an open-source Python framework for teams that need to analyze data or train machine learning models across separately held datasets. It brings several privacy-preserving computing methods into one framework, so applications can combine them according to their security assumptions and performance needs.

Aftershoot is a photo processing app for professional photographers who need to review and edit whole shoots. Its AI culling, editing and retouching run locally on Windows and macOS, without uploading images for processing.

MLX Audio is a Python library for developers building speech and audio applications on Apple Silicon Macs. It runs models locally using Apple's MLX framework, with support for generated speech, transcription and audio processing.

pyannote.audio is a Python toolkit for developers and researchers who need to separate recordings by speaker. Speaker diarization identifies who spoke when, so applications can associate parts of an audio recording with different speakers.

Accentize dxSplit is an AI audio plugin for dialogue editors and post-production mixers who need to clean up speech recordings. It separates dialogue into voice, room reverb and background noise so you can adjust each component independently.

ToolJet is an internal-app builder for developers and operations teams creating dashboards, admin panels and business tools around existing data. It can run on your own infrastructure with Docker or Kubernetes, or through ToolJet Cloud. The community foundation uses AGPL-3.0; Enterprise features use separate commercial terms.

jetson-stats monitors NVIDIA Jetson boards and controls their power and cooling settings. It's for developers running workloads, including local AI, who need to inspect resource use and adjust how their board operates.

FableCut is a local video editor for people who want to make cuts themselves and work alongside an AI agent. Both can edit the same project, with agent changes appearing live in the browser.

ncnn is a C++ inference framework for developers adding on-device AI to mobile, desktop and embedded applications. It focuses on a small memory footprint and doesn't require third-party runtime dependencies.

DVC is a data version control tool for individual data scientists and small machine learning projects. It links datasets and model artifacts to code history so you can recover the inputs behind an experiment and reproduce its results.

Torrix is a self-hosted LLM monitoring tool for teams investigating AI spending and failed requests. It keeps request records in your infrastructure and brings usage analytics and individual traces into one dashboard.

Depth Pro estimates distances from a single image for researchers and developers who need depth maps with an absolute scale. It can make predictions without camera metadata, so images don't need to come with known camera intrinsics.

Agent QA is a source-available AI testing agent for developers and software teams testing web, Android and iOS apps. You describe a user journey and its expected results in plain English. It tests the real interface and carries observations from earlier runs into later ones, so it can account for how your app behaves.

Jina Embeddings is a family of models for developers building semantic search, retrieval-augmented generation (RAG) and AI agent applications. The models convert content into vectors that a search system can compare for relevance.

InternVL is a family of vision-language models for developers and researchers building AI applications that interpret images and answer questions about them. Downloadable models let teams host inference themselves.

Interpreter translates Japanese game text into English subtitles over the game window. It's aimed at retro-game players, with MeikiOCR tuned for Japanese game text and pixel fonts. The default MeikiOCR and Sugoi V4 models run locally without internet access after setup.

SwarmVault is a self-hosted knowledge base for people organizing research, notes or code documentation for themselves and AI agents. It turns collected sources into a persistent Markdown wiki with connections you can inspect.

Forge Orchestrator coordinates Claude Code, Codex CLI and Gemini CLI working on the same repository, so they can share decisions and avoid conflicting edits. It's built for internal use at NXTG; outside use isn't supported, there's no SLA, and interfaces may change without notice.

HunyuanWorld generates explorable 3D environments from text prompts or images. It's for game developers, VR creators and researchers who need scene geometry for interactive content or physical simulation.

Playwright MCP gives AI agents browser control for testing and tasks that involve interacting with web pages. It reads structured accessibility snapshots, so agents can identify page elements without a vision model.

MEX keeps project knowledge alongside code so engineers and their AI coding agents can share context across teammates and sessions. Architecture, requirements and decisions live in readable Markdown that travels with the repository through Git.

FastMCP is a Python framework for developers who want AI agents to use their applications and data through the Model Context Protocol (MCP). It covers both publishing tools and connecting to them, with support for servers that run locally or remotely.

Subtitle Edit is an offline desktop application for people creating captions or revising subtitle files alongside video. Editing, format conversion and playback run on your computer. Optional translation and speech services send requests to external providers.

PentestGPT is an AI agent for security testers working on authorized penetration tests and capture-the-flag (CTF) challenges. It can carry out a test autonomously or let you guide the work in an interactive mode.

llamafile packages a language model and its runtime in a single executable for people who want to run models locally or share a ready-to-run model with others. Recipients don't need to install a separate runtime.

APIPark is an AI gateway and API developer portal for teams managing model services and business APIs. It gives applications a common request format and lets teams share APIs with controlled access.

Accentize dxRevive is a proprietary AI speech restoration plugin for macOS and Windows that reconstructs missing frequency content as well as reducing noise. It's aimed at podcast editors, content creators and dialogue editors working with damaged or poorly recorded speech. All audio processing stays on your computer, without cloud uploads or special hardware.

nvitop is a terminal GPU monitor for people running AI workloads on NVIDIA hardware. It shows GPU activity alongside the processes using it, so you can check resource use and investigate busy devices.

Kition is a desktop AI workspace for writing, research, and project planning. Its agent works directly with editable project files, so you can inspect proposed changes in the documents and boards they affect.

AutoAgent lets people create AI agents by describing the work they want done. It's for users who want custom agents without writing the agent code themselves, or a ready-made research assistant for gathering information and producing reports.

HiKid is a desktop app for children in non-English-speaking countries to practice speaking and listening. Kids talk with an AI in English through a microphone, without needing to type.

Lotti is a private logbook for people who want to keep their plans and a record of what actually happened together. It connects daily planning with personal notes and work history while keeping the underlying record on your devices.

Codeg is an AI coding workspace for developers who use several agents and want to coordinate their work without losing context between tools.

Context7 is a locally runnable MCP server and CLI that connect coding assistants to its hosted documentation service. It's for developers who want library references and code examples in their assistant's context. The local component is the connection to that service: the API backend, document parsing and crawling engines remain private hosted components.

SemanticFinder helps readers and researchers find passages by meaning in long documents. It runs the search model in your browser, so you can search your own text without sending it to a model server.