𓁟

Local-First AI Assistant

Your AI runs here.
Your data stays here.

Thoth is a desktop AI assistant with memory, tools, workflows, design creation, messaging, plugins, MCP tools, and provider-aware model routing. Run it fully local with Ollama, add frontier APIs, or sign in to ChatGPT / Codex when you want subscription-backed models. Your durable data stays on your machine.

Download for Windows .exe Download for macOS .dmg Install on Linux curl View on GitHub
0 core tools 0 bundled skills 0 local models Apache 2.0 · Free
See it in action - a quick walkthrough of the core experience
Thoth is built for personal AI sovereignty: your compute, your data, your model routes, and your automation under your control.

The strongest pieces first

Designer Studio

Create decks, documents, landing pages, app mockups, and storyboards with live preview, surgical AI edits, media generation, chart insertion, export, and shareable interactive links.

Five modes

Personal Knowledge Graph

Memories become connected entities and typed relationships, with semantic recall, graph expansion, visual exploration, document extraction, and Obsidian-compatible wiki export.

Durable memory

Real Computer Use

The agent can browse in Chromium, use the shell, manage files, search the web, read documents, send email, manage calendar events, inspect status, and create charts.

30 core tools

Workflows & Approvals

Run scheduled pipelines with conditions, approvals, subtasks, webhooks, notifications, model/tool overrides, safety modes, run history, and insight cards.

Automation

Channels, Voice & Media

Telegram, WhatsApp, Discord, Slack, and SMS share streaming, media intake, reactions, approvals, health checks, local speech-to-text, and Kokoro TTS.

5 channels

Plugins, MCP & Delegation

Install sandboxed plugins, connect external MCP servers, coordinate Claude Code CLI through an approval-gated skill, and migrate selected Hermes/OpenClaw data with backups.

Extensible

Image & Video Generation

Generate or edit images through OpenAI, Google, and xAI, then create video clips through Google Veo or xAI Grok Imagine Video from the same provider-aware model catalog.

Provider media

Provider Runtime & Models

OpenAI, Anthropic, Google AI, xAI, OpenRouter, MiniMax, custom endpoints, local Ollama rows, and ChatGPT / Codex all flow through one catalog with provider-aware labels.

Updated in 3.20

System maps

Explore the architecture behind local-first control.

The visual architecture gallery shows how Thoth connects its agent orchestrator, context assembly, memory graph, background workflows, channels, Designer Studio, safety boundaries, and self-awareness layer without giving up user control.

8 subsystem diagrams
Local-first data boundaries
Tool and provider routing
Safety and approval flows
View architecture
Thoth core agent architecture preview
Core Agent Architecture plus seven subsystem maps

Memory that compounds

Thoth remembers people, projects, preferences, facts, events, and relationships as a graph, then retrieves relevant nodes and one-hop context before each response.

Document extraction, Dream Cycle refinement, provider-aware insight snapshots, wiki export, and the interactive graph view keep that memory useful instead of turning it into a pile of notes.

Local when you can. Providers when you want.

One assistant, local models, API providers, custom endpoints, MiniMax, and ChatGPT / Codex

Local models · Ollama

Private by default

Run inference on your hardware with 39 curated models across Qwen, Llama, Gemma, Mistral, DeepSeek, Phi, and more.

  • Works offline after model download
  • No subscription or provider account
  • Default local model: qwen3:14b
Providers & subscriptions · Opt-in

Frontier reasoning on demand

Add OpenAI, Anthropic, Google AI, xAI, OpenRouter, MiniMax, custom OpenAI-compatible endpoints, or ChatGPT / Codex when you need frontier reasoning or do not have a GPU.

  • Settings → Providers handles sign-in, health, and keys
  • Settings → Models handles catalogs, defaults, and Quick Choices
  • Duplicate names are labeled by route: API, custom endpoint, or ChatGPT / Codex

30 core tools the agent wields for you

Ask a question in natural language and the agent autonomously decides which tools to chain together. Search the web, check your email, run a shell command, inspect provider and insight status, build a deck in Designer, and chart the results - all in a single conversation turn.

Connected channels add send tools automatically, external MCP servers add dynamic tools, and every risky action stays behind Thoth's approval gates.

30 Core Tools + Channel Tools

Core tools ship with the app; channels and MCP servers extend the tool belt when enabled

Search & Knowledge × 8
Web Search API
DuckDuckGo Free
Wikipedia Free
arXiv Free
YouTube Free
URL Reader Free
Documents Free
Wiki Vault Free
Productivity × 8
Gmail OAuth
Calendar OAuth
Filesystem Free
Shell Free
Browser Free
Workflows Free
Tracker Free
X (Twitter) OAuth
Computation, Memory & Media × 9
Calculator Free
Wolfram Alpha API
Weather Free
Vision Free
Memory Free
Conversation Search Free
Charts Free
Image Gen API
Video Gen API
Design, Platform & External Tools × 5
Designer Free
Thoth Status Free
MCP Tools Free
Updater Free
System Info Free

Providers, workflows, channels, and external tools

Configure provider accounts, pin model Quick Choices, build presentations and interactive pages, automate recurring work, route approvals to messaging channels, and connect MCP servers without making startup fragile.

ChatGPT / Codex sign-in is separate from OpenAI API keys, Claude Code Delegation stays approval-gated, and imports from Hermes/OpenClaw remain scan-first with backups and redacted reports.

Why not just use cloud chat?

Cloud AI rents you access. Thoth gives you ownership.

Cloud AI 𓁟 Thoth
Your data Stored on provider servers, subject to their policies Stays on your machine - always. Provider models receive only the current conversation plus model-visible tool context
Conversations Provider-owned - can be deleted, leaked, or used for training Local SQLite, fully yours, exportable anytime
Cost $20+/month subscription Free with local models. Providers use upstream API billing or your ChatGPT subscription
Memory Limited, opaque, provider-controlled Knowledge graph - 10 entity types, 67 relation types, visual explorer, FAISS semantic search
Tools Sandboxed plugins, limited integrations 30 core tools + channels + plugins + MCP tools
Customisation Pick a model, write a system prompt Swap models per thread, mix local + providers, configure identity, enable skills and guides, build workflows, install plugins
Voice Cloud-processed speech Local Whisper STT + Kokoro TTS - never leaves your mic
Messaging No messaging integrations 5 channels - Telegram, WhatsApp, Discord, Slack, SMS - with streaming and media
Availability Requires internet, subject to outages Local models work fully offline

Up and running in minutes

↓ 01

Download & install

Use the one-click Windows installer, macOS DMG, or verified Linux one-line installer. Each release ships self-contained runtime assets for the supported platform.

↔ 02

Choose your path

First-launch wizard: Local - download an Ollama model (default: qwen3:14b), or Providers - enter an API key and seed Quick Choices. ChatGPT / Codex sign-in lives in Settings after launch.

▶ 03

Start talking

Ask anything in natural language - the agent handles the rest. Search the web, check email, analyze your screen, track a medication, schedule a workflow.

Linux install

Installs the latest verified XDG user tarball, creates ~/.local/bin/thoth, and uses the same release manifest as in-app updates. Manual tarballs remain available from GitHub Releases.

curl -fsSL https://raw.githubusercontent.com/siddsachar/Thoth/main/installer/install-linux.sh | bash

Privacy by architecture

Thoth's app has no account system, server, or telemetry pipeline. Conversations, memory, Designer projects, documents, and local model runs stay on your machine. Provider models, ChatGPT / Codex, and online tools are opt-in, with secrets stored in the OS credential store when available.

◈ Local inference (default) ☁ Providers opt-in only ■ Local storage ◯ Local voice ✕ No app telemetry → Opt-in APIs ◊ Injection defence ⚖ Apache 2.0

What’s new in v3.20

Read the full release notes →

Take back your AI sovereignty

Free, open source, and yours to keep. No Thoth account. No Thoth subscription. No data harvesting.

Built with
NiceGUILangGraphLangChain OllamaFAISSKokoro TTS pywebviewfaster-whisper