Vellum

Open-source personal AI assistant with long-term memory that automates tasks across your tools and files.

Vellum provides an open-source personal AI assistant designed to handle "busywork" by learning user preferences and executing tasks across various digital tools. Unlike standard chatbots, Vellum is an autonomous agent with long-term memory that can take real actions like sending emails, managing calendars, and updating project management boards. It is built on the principle that personal data should remain private, offering local execution on a user's machine or isolated containers in the cloud.

The platform is designed for both personal and business use, functioning as a "second brain" or an "AI employee." It integrates with popular tools such as Google Workspace, Slack, Notion, GitHub, and Linear. Users can customize their assistant's identity and personality, and extend its capabilities through a library of community-built plugins or custom skills.

Vellum is native to macOS but accessible across iOS, Android, web browsers, and messaging platforms like Telegram and Slack. While the company announced the shutdown of its hosted cloud platform in late 2026, the core assistant remains open-source and available for self-hosting, allowing users to maintain their assistants indefinitely on their own hardware.

Founded
2024

Industries: Artificial Intelligence, SaaS

Categories: AI Agents, Productivity Software

AntSpace put this profile together from public information. It may be incomplete; once Vellum claims it, its admins can edit everything here.

Our mission

Vellum's mission is to build the most intelligent personal layer and bring personal intelligence to everyone, focusing on security, privacy, and avoiding model lock-in.

Our vision

Vellum envisions a world where every person has a personal AI assistant that learns their preferences, remembers their life, and handles tasks across their tools automatically.

Our culture

Vellum describes itself as a small team obsessed with giving everyone a brilliant AI assistant. The company operates under a set of core principles: AI shouldn't feel like work, simplicity by design, and being fun on purpose. They aim to make setting up an assistant feel like meeting someone new rather than filling out a form, revealing capabilities only as they become relevant to avoid overwhelming users. The team values leaning into the future, saying what they mean, and playing to win. Every person at the company uses their own AI assistant in their daily work.

Core values

  • Lean into the future
  • Say what you mean; mean what you say
  • Play to win
  • AI shouldn't feel like work
  • Simple by design
  • Fun on purpose

Specialties

  • Personal AI Assistants
  • Open Source AI

Markets served

  • SaaS

Latest from Vellum

  • v0.12.9 Release (Product launch · November 8, 2026)
    Vellum released version 0.12.9 of its assistant, introducing a fix for workspace restorations. The update ensures that full Git history is preserved when restoring from backups, even for…
  • Vellum’s Personal Assistant Platform is shutting down (Service announcement · November 7, 2026)
    Vellum announced the closure of its hosted Personal Assistant Platform, effective December 13, 2026. The company will provide final data exports for cloud-hosted assistants until that date, while…

All 2 updates

Products by Vellum

  • Vellum Platform (SaaS / cloud software · From 30 USD month)
    Build AI agents for your boring ops tasks with no code required.

All 4 products

Services by Vellum

  • Vellum Cloud Hosting (Managed service · From 0 USD)
    Managed cloud infrastructure for personal AI assistants with dedicated compute and storage.

All 1 services

Clients and partners

4 clients and partners

All 4 clients and partners

Case studies

2 case studies

All 2 case studies

Documents

2 documents

All 2 documents

3 more products, services and AI on AntSpace. Sign up free to see the full catalog.

Leadership, clients, case studies, documents and every location on AntSpace. Sign up free to see them.

Related