Local & Self-Hosted AI Tools

Mirascope is an open-source development library for building AI applications with local models or cloud providers. Its documented local backends include Ollama, vLLM and LM Studio, plus direct inference through MLX on Apple Silicon Macs. Developers can use a shared model interface while keeping inference on their own hardware or sending requests to OpenAI, Anthropic or Google.

mistral.rs is an inference engine for developers who want to run AI models on their own computer or server. It supports browser chat and inference embedded directly in Python or Rust applications.

DiffSinger generates synthetic singing voices for voice developers and music production. OpenVPI maintains this fork of MoonInTheRiver/DiffSinger, with acoustic models and adjustable vocal expression controls.

Chrome DevTools MCP connects AI coding assistants to a live Chrome browser through a locally running server. It's for developers who want an agent to investigate web app behavior alongside its coding work.

npcpy is a Python library for developers and researchers building AI agents and applications with language models. It supports local LLM backends as well as cloud providers, so projects can choose where model requests run.

HanaAgent is an open-source desktop AI agent for everyday computer work and coding, with user data stored on your machine. It runs on macOS, including Apple Silicon and Intel, and Linux. It supports Ollama local models as well as OpenAI-compatible, Anthropic-style and OAuth provider connections. Local inference uses Ollama; requests to a cloud provider go to that service.

Kokoro is a text-to-speech model and inference library for developers adding speech generation to personal projects or production applications. It runs on your own hardware and uses the compact Kokoro-82M model.

tlm is a terminal assistant for people who want help writing shell commands or understanding commands they've found elsewhere. It uses local models through Ollama rather than a cloud AI service.

AudioBud is a dictation app for Windows that puts speech into the text field you're writing in. It builds on Handy by CJ Pais and suits people who want voice typing with local speech recognition rather than a hosted transcription service.

Headroom reduces the context that LLM apps and coding agents send to a model. It's for developers whose prompts fill up with repetitive tool results, logs, file contents and retrieved documents. Compression runs locally, but model requests still go to the chosen LLM provider.

PR-Agent is an open-source AI code review agent for development teams that want to run their own pull request reviewer. It can provide feedback automatically through repository workflows or respond to requests in pull request comments.

Topaz Video is a video processing app for people who want to work on footage using their own hardware. It combines a choice of video models with stabilization and SDR-to-HDR conversion.

Uni MoE is a family of multimodal AI models for researchers working with language, visual material and audio together. Its Omni model can reason across combined inputs and answer questions about their content.

Lance is an open source data format for teams building search engines, feature stores and machine learning training pipelines on object storage. It keeps images, video, audio, text and embeddings alongside tabular data, so those workloads can share a dataset. The project uses the Apache 2.0 license.

LLM Sandbox is a Python library for developers whose AI agents or applications need to execute generated code. It puts execution in containers on infrastructure you control, with support for local deployments and Kubernetes clusters.

Antinode Clear is a proprietary audio plug-in for Windows and macOS that separates speech from background noise and voice reverb in real time. Formerly called Supertone Clear, it's for post-production engineers, video editors, musicians and streamers who need to clean up speech or retain the ambience behind it.

Jamba is AI21's language model family for teams building document-based assistants and business applications on their own servers. The Large model targets lengthy context, but its hardware requirements put it in server territory: quantized deployment uses eight GPUs with 80GB of memory each.

HiDream-I1 generates images from text prompts for creators and developers who want to work on their own hardware or add image generation to an application.

DwarfStar is a local LLM inference engine for people who want to run large DeepSeek, GLM and Qwen models on their own computers or servers. It focuses on selected models and hardware setups, and requires the project's own GGUF files rather than accepting arbitrary GGUF downloads.

Luminar Neo is a photo editor for photographers who want automatic adjustments alongside manual control over their images. It combines tools for portraits and outdoor scenes with layers, masks and edits you can revise without altering the original.

nono is a security runtime for people running terminal AI agents on their own machines or servers. It uses operating system boundaries to restrict an agent's access to files, network services and credentials.

Bitterbot is a personal AI agent for people who want an assistant that remembers ongoing work across conversations. It runs on your own machine and can keep working between conversations while the machine is on.

Spring AI is a Java framework for developers adding AI features to Spring applications. It connects models with application data and APIs for tasks such as answering questions about documents.

Open Glean is a self-hosted AI workspace for searching files, notes, saved webpages and knowledge synced through Hydra DB connectors. Its browser app runs on your machine or server, including in Docker, but retrieval uses the remote Hydra DB API by default and requires a Hydra key. You can change the Hydra backend endpoint. Local answer models require explicit enablement and are disabled by default.

ReScript is a video and audio editor for people cutting interviews, podcasts and voice notes through text. Deleting words from a transcript removes the matching sections of the recording.

Modly is a desktop app for artists, game developers and people preparing models for 3D printing. It turns a photo or text prompt into a textured 3D mesh using AI models on your own GPU.

OpenDesign is a design workspace for teams that want to turn briefs into working prototypes and visual deliverables with their chosen coding agent. It keeps previews and editable project files together so designers and developers can refine the same work.

Mono Color Skill helps designers and publishers turn a theme, phrase, object or photograph into editorial artwork for posters, zines, covers and packaging. It focuses on limited ink palettes, visible paper and print textures.

Draw Things creates and edits images on iPhone, iPad and Mac. It's for artists and other creators who want to work offline and keep generation on their own hardware.

Labaik is a desktop AI workspace for people who use chat models for writing, research and coding. It brings local and cloud models into the same conversation interface, with model switching during a chat.

gp.nvim brings AI chat and prompt-based editing into Neovim for developers and writers who want to work with their existing buffers. You can ask about selected text or code, then apply the response inside the editor.

Odysseus brings AI chat, autonomous agents and personal work into one workspace on your own hardware. It's for people who want models to work with their documents and ongoing tasks in a self-hosted app.

Computer Use Linux is a local MCP server and CLI for AI agents that need to interact with Linux desktop apps. Agents can identify accessible controls by role, name or text instead of relying entirely on screenshot coordinates.

Tabby is a self-hosted AI coding assistant for developers and teams who want a GitHub Copilot alternative on their own infrastructure. It brings coding suggestions and answers into the editor.

BambooAI is a self-hosted AI data analyst for people working with tabular datasets who want to inspect the calculations behind an answer. It turns plain-language questions into Python analysis and charts, then reruns the calculations cited in its report against the original data to check that the numbers reproduce.

LitGPT is a Python toolkit for developers and researchers who want to train and deploy language models on their own hardware. Its directly editable model implementations suit custom applications and research experiments.

Transcribe Anything turns recordings and online videos into text and subtitles using Whisper-based speech recognition. It's a command-line tool with a Python API for people processing media locally or adding transcription to their own applications.

SurfSense is a desktop NotebookLM alternative for people working with confidential documents, research or study material. It answers questions across your files and turns selected sources into editable outputs.

AgenticSeek is a local alternative to Manus AI for people who want an assistant to carry out research, coding and file tasks on their own hardware. It can break a request into steps and assign the work to specialized agents.

SmolChat is an Android app for chatting with language models on your phone. It gives people who want on-device AI a chat interface for their own GGUF model files, including small language models.

Claw-Empire is a self-hosted AI agent manager for people who want to delegate projects to multiple coding assistants. Its pixel-art office shows agents working across departments, while a shared dashboard tracks their tasks.

Ouroboros is a locally installed workflow engine for developers who want to define requirements and check an AI coding agent's output. It runs through a terminal CLI, a Claude Code plugin or an MCP server. Your chosen coding runtime supplies the agent and model access. When used with Claude through Claude Code's standard cloud connection, model requests go to that external service.

Open Wispr is a voice dictation tool for people who want to speak into Mac apps while keeping audio and text on their computer. It inserts transcribed speech at the cursor in editors, browsers, terminals, email and other applications.

OpenSuperWhisper is a speech-to-text app for people who want to dictate or transcribe saved recordings on an Apple Silicon Mac. It runs Whisper and Parakeet models locally.

Memex is an AI journal for people who capture scattered notes, photos, and voice recordings and want help organizing them. It turns those fragments into structured records and connects information across entries.

Claude Context adds repository search to AI coding assistants so they can retrieve relevant code without loading whole directories into their context. It's for developers working with large codebases that an assistant needs to understand.