Open source · AGPL-3.0 · Self-hosted

An AI workspace that answers to no one but you.

Zenith runs a full AI workspace — chat, a coding agent, deep research, long-term memory, image generation, voice — entirely on your own machine, against your own Ollama models. Nothing you type is ever sent anywhere.

zenith · session
endpoint127.0.0.1
requests leaving this machine0
account requirednone
monthly cost$0
status● running locally

Why Zenith

Most AI chat apps run on someone else's computer.

Every message, every file, every question — sent to a server you can't see, kept under terms you didn't write. Zenith doesn't work that way.

It runs entirely on hardware you control, talking only to Ollama — a free, local model runner — on the same machine, or your own network. No account. No subscription. No API key, ever.

Cloud assistants send everything you type to a remote server
Your data trains models or gets logged, per someone else's policy
A subscription, a rate limit, or a paywall for real use
+Zenith runs entirely on your own hardware — nothing leaves
+Open source under AGPL-3.0, forever free

Feature tour

Everything a workspace should have

Not a chatbot demo — a daily-driver, verified against a real running deployment.

Chat & multimodal

Streaming chat, images, documents, video — routed automatically to whichever installed model fits the message.

Coding agent

Shell, file edits with diff previews, persistent sessions, real git — with an approval gate from manual to full-auto, your call.

Deep research

Multi-step web research that plans, searches, and writes a cited report — refuses to answer before it's actually searched.

Long-term memory

Learns durable facts about you and recalls them automatically — with a conflict review for things that no longer add up.

Code editor + git

A real file tree, editor, and git status scoped to a working directory — extract a conversation's code into a runnable project in one click.

Knowledge graph

See how conversations, memories, and documents actually connect — built from real relationships, not a guess.

Automation & webhooks

Trigger → action rules on real backend events, plus outbound webhooks so your own scripts can react to what Zenith is doing.

Voice in & out

Hold the mic to talk, hear replies spoken back — including a hands-free continuous conversation mode.

Image generation

Bring your own local Stable Diffusion server — generated images render right in the chat.

MCP tool servers

Connect external Model Context Protocol servers instead of waiting for a built-in integration — a curated starter list included.

Passcode lock & encryption

Real API-level lock middleware for shared machines, plus AES-256-GCM encrypted credentials at rest.

Desktop app

A real installable app with a system tray, global hotkey, and auto-update — not just a browser tab.

Who it's for

Same app, very different reasons to open it

Zenith has a lot of surface area (see the feature tour above). Nobody uses all of it. Find yourself below and turn on just that subset.

"I just want a private ChatGPT"

You want everyday conversation — typed or spoken — that isn't training a model somewhere else. Nothing fancier than that.

  • Chat & multimodal — works the moment you install, no setup
  • Voice in & out — talk instead of type, hear replies back
  • Long-term memory — it remembers who you are between chats
  • Image generation — pictures, without a separate app

Skip: agent mode, MCP, the code editor — none of that shows up unless you go looking for it.

"I'm a developer"

You want a coding agent with real teeth — one that can touch your filesystem and git history, not just paste snippets into a chat box.

  • Coding agent — shell + file edits + git, with an approval gate you control
  • Code editor + git — a real file tree and diff view scoped to your working directory
  • MCP tool servers — bring in tools Zenith doesn't ship natively
  • Deep research — cited answers for API/library lookups instead of guesses
  • Knowledge graph — a memory layer that tracks how a codebase-adjacent project connects

"I'm a writer, researcher, or student"

You're working from sources — papers, drafts, a growing pile of context — and need the app to hold that across sessions, not just one chat.

  • Deep research — a cited, reopenable report instead of an unsourced answer
  • Document RAG — upload PDFs/papers and ask questions grounded in them
  • Long-term memory — keeps facts about an ongoing project straight
  • Conversation branching — explore an alternate draft without losing the original thread

"Privacy-conscious first, features second"

You picked Zenith specifically because nothing leaves your machine by default — and you want to know exactly where the few exceptions are before you flip anything on.

  • Passcode lock & encryption — real API-level lock, AES-256-GCM credentials at rest
  • Single-user, no telemetry — no account, no analytics, nothing phoned home
  • The deliberate exceptions — web search, reading a URL you gave it, and optional Ollama Cloud models are the only features that talk to the outside world, and only when you explicitly turn them on for that message

"I like to tinker / self-host everything"

You want the whole instrument panel — rules that fire on real events, tools wired in from elsewhere, and access from more than one device.

  • Automation & webhooks — trigger → action rules, plus outbound webhooks for your own scripts
  • MCP tool servers — connect anything speaking Model Context Protocol
  • Folder watching & scheduled turns — react to files changing, run recurring prompts unattended
  • Task board — the Keep/Calendar/To-do suite as a lightweight ops board
  • Desktop app + multi-device sync — tray icon on one machine, paired access from another

Get started

Three ways to run it

Pick whichever matches how comfortable you are with a terminal.

Desktop app

double-click · no terminal
  • You want a normal app icon, tray, and hotkey
  • You'd rather not touch a command line at all
  • Windows .msi / Linux .AppImage, built by CI
Desktop app guide →

Manual

full control · no Docker
  • You want the backend and frontend running directly
  • You're already comfortable with Python/Node
  • One script checks prerequisites and sets everything up
Manual setup →

Before you install

What hardware do you actually need?

Zenith itself is lightweight — the real requirement is whatever Ollama model you plan to run. Settings → Model routing scores your actual installed models against your real GPU/RAM automatically; this is the general shape of it.

Your GPU VRAM (or RAM, CPU-only)Comfortable model sizeStill usable
No dedicated GPU / under 4 GB1–2B paramsup to ~3B
4–8 GB (e.g. RTX 3050/3060 laptop)3–4B paramsup to ~8B
8–16 GB (e.g. RTX 4060 Ti/4070)8–12B paramsup to ~20B
16 GB+ (e.g. RTX 4090, workstation)20B+ paramswhatever fits

Rough heuristic (~0.65GB per billion params at Q4 quantization, plus runtime overhead) — not a promise. A model too big for your GPU doesn't error, it just spills onto much slower CPU inference; Zenith's own stuck-generation watchdog catches and reports this instead of leaving you staring at a spinner.

FAQ

Common questions

Is it actually free?+

Yes — Zenith is open source under AGPL-3.0, no subscription, no paid tier, no usage limits. The only ongoing cost is your own computer's electricity. Optional features that touch the open web (search, page reading) use free, keyless endpoints, not paid APIs.

Does any of my data leave my computer?+

No, by default. Zenith only talks to your own local Ollama instance and, only when a feature that needs it is explicitly turned on for that message, the open web (search results, a page you asked it to read). There's no telemetry, no analytics, no account.

Do I need a powerful GPU?+

No — it runs on CPU-only setups too, just slower. See the hardware guide above; even modest laptop GPUs run small models comfortably.

Can I use it on my phone?+

Yes — it's an installable PWA with a mobile-friendly layout. Open it from your phone's browser over your LAN or Tailscale and add it to your home screen.

What's the license, in plain terms?+

AGPL-3.0-or-later: use, modify, and self-host it freely. If you run a modified version as a network service others can use, you have to publish your changes too — a deliberate choice, matching the project's "no cloud middleman" philosophy.

Is this affiliated with Ollama?+

No — Zenith is an independent, unaffiliated client that talks to Ollama's own local API, the same way any other app on your machine would.