Skip to content

Category: Ember Desktop · AI Work Operating System

Your AI work stays on your machine — organized, persistent, yours.

Ember Desktop is a local-first workspace for conversation, deep research, coding agents and parallel work. Conversations, files and artifacts live in durable workspaces on your machine — not in disposable cloud chat threads. In active client testing on macOS and Windows.

Built for

  • Knowledge workers
  • Developers
  • Research teams
Status
Beta
Ledger step
Ledger step 3 of 4
Family
Ember family
Access
This site only
Listed
3 / 4
Active client testing
Beta
Native desktop apps
macOS + Windows
Lightweight native shell
Tauri 2 / Rust
Data stays on-device
Local-first

What this page can show today

  • The register marks this track as a public test or verification build.
  • This track is documented on this site only.
  • Capabilities below are the published feature list for this track.

Limits of this page

  • No public product site is published for this track.
  • Beta is not a production claim.
  • This page does not publish financials, headcount, or client names.

Key features

  1. 01

    Persistent workspaces

    Conversations, files, tasks, traces and artifacts remain organized around durable workspaces instead of disposable chat threads.

  2. 02

    Parallel, evidence-gated agents

    Plan-and-act agents can work in parallel with file-level coordination, permission modes and visible execution traces.

  3. 03

    Local-first control

    Provider credentials live in the operating-system keychain, local models are supported, and tool execution is permission-gated.

  4. 04

    Research and artifact studio

    Deep-research reports, code, documents, charts, tables and generated images can be saved, versioned and exported.

Published links

Technology stack

Tauri 2 · Rust · React · TypeScript · SQLite · Ember Gateway

Related platforms

Family on the register: Ember family Same family first, then tracks at the same lifecycle stage.

  • Same family · Live

    Ember AI

    Foundation Models

    Public product site

  • Same family · Live

    Ember Legal

    Legal Technology

    Invite only

  • Same family · Beta

    Ember Live

    Real-Time Model Surface

    Invite only

  • Same family · Beta

    Ember Agent OS

    Multi-Agent Operations

    This site only

This cell on the investor register

Status and family write into the /investors URL. The holdings table lists every public track in that cell.

Open Beta Ember family holdings
3 / 4

JSON fields are copied from this row in src/data/platforms.ts. No financials.

How this track works

Audience on the left, the published Ember stack in the middle, capabilities on the right. Every node is copied from this track in the public register.

How this track moves work from its published audience through Coreidea and Ember into listed capabilities.

Inputs

Coreidea and Ember

Outputs

Select a node to inspect the published role and capability.

Cloud AI tools treat your work as disposable.

Browser-based AI assistants lose context between sessions, store your data on someone else's servers, and collapse complex, multi-day work into a single scrolling thread.

  • Context dies with the tab

    Close the browser and the conversation is gone. Multi-day research, coding sessions and analysis have no durable home — every session starts from zero.

  • Your data lives on their servers

    Cloud-first assistants store prompts, files and outputs remotely. For sensitive business, legal or research work, that is a trust boundary most teams never wanted to cross.

  • Real work outgrows one thread

    Real projects span parallel tasks, multiple files, code, charts and documents. A single chat thread flattens all of it into an unsearchable scroll.

A workspace, not a chat window.

Ember Desktop was designed around one conviction: serious AI work deserves the same durability, organization and privacy as any other professional tool.

  • Persistent workspaces

    Conversations, files, tasks, traces and artifacts stay organized around durable workspaces. Close the app, reopen next week — everything is where you left it.

  • Parallel, evidence-gated agents

    Plan-and-act agents work in parallel with file-level coordination, permission modes and visible execution traces — you see what each agent did and why.

  • Credentials in the OS keychain

    Provider API keys live in the operating-system keychain, never in plaintext config files. Local models are supported alongside cloud providers.

  • No telemetry by default

    No telemetry, no cloud sync unless you explicitly enable it. What you build in a workspace stays in that workspace.

Install. Open a workspace. Get real work done.

  1. Install the native app

    Download the macOS or Windows app. The Tauri 2 / Rust shell is lightweight — no browser, no Electron bloat.

  2. Create a workspace

    Name a workspace for the project or topic. It becomes the durable container for conversations, files, agents and artifacts.

  3. Connect providers via keychain

    Add API keys through the OS keychain prompt, or point at a local model. No plaintext config files.

  4. Run conversations and agents

    Chat, run deep-research reports, launch coding agents in parallel. Permission gates control what each agent can touch.

  5. Save, version and export artifacts

    Reports, code, documents, charts and images are saved as artifacts inside the workspace — versioned and exportable.

Built for trust on your machine.

  • Tauri 2 + Rust shell

    A lightweight native shell with a React + TypeScript front end and SQLite persistence. Small binary, fast start, no bundled browser engine.

  • Keychain-native credentials

    API keys are stored in the operating-system keychain — macOS Keychain or Windows Credential Manager — never in editable config files.

  • Permission-gated execution

    Tool execution is sandboxed and permission-gated. Agents request access; you approve or deny. Every action leaves a visible trace.

  • Ember Gateway integration

    Optionally route inference through the Coreidea private gateway running the Ember AI model family, or use your own provider keys and local models.

Who Ember Desktop is built for.

  • Knowledge workers

    Deep-research reports, document analysis and multi-source synthesis — organized in workspaces that survive across sessions.

  • Developers

    Coding agents with file-level coordination, visible execution traces and permission modes — parallel work without losing the thread.

  • Research teams

    Save, version and export research artifacts. Workspaces keep the full trail — conversations, data, code and outputs — in one place.

  • Privacy-conscious professionals

    No telemetry, no cloud sync by default, credentials in the OS keychain. Sensitive work stays on your machine.

Questions, answered.

What stage is Ember Desktop at?

Ember Desktop is in active client testing on macOS and Windows. It is not yet a public web service. The core workspace, agent and artifact workflows are functional; we are hardening edge cases and gathering feedback from testing partners.

Does my data leave my machine?

By default, no. Workspaces, conversations and artifacts are stored locally in SQLite. Cloud inference is only used when you connect a provider; no telemetry or sync runs unless you explicitly enable it.

Can I use local models?

Yes. Ember Desktop supports local models alongside cloud providers. You can also route through the Coreidea private gateway running the Ember AI model family.

How do I get access?

Ember Desktop is in active client testing. Contact info@coreidea.cn to request access or learn about the testing program.

AI work that respects your data and stays organized.

Ember Desktop is in active client testing on macOS and Windows. Request access, or talk to us about your workflow.

All platforms