Skip to main content
Back to discovery
🧩
Free plan + paidInformation reviewed 2026-10-15

Vellum

Visit Website

About this listing

Product and access model reviewed from current official sources; workflow and example are editorial suggestions, not a hands-on test.

Official source

A personal AI assistant with persistent memory, connected tools, scheduled work, and hosted or self-hosted operation.

Best suited to

Vellum's current offering is a personal assistant that can retain preferences and perform tasks across connected tools.

Meet Vellum

Vellum's current offering is a personal assistant that can retain preferences and perform tasks across connected tools. It suits users who want continuity between conversations and repeated work. Start with a read-only task before enabling actions. Memory can retain an incorrect assumption as well as a useful fact, so inspect important context and set clear permission boundaries for messages, bookings and other outside changes.

A practical workflow

  1. Choose hosted access or review self-hosting requirements.
  2. Configure permitted sources and action boundaries.
  3. Ask for a small read-only result.
  4. Verify remembered facts and outputs before enabling automation.

Illustrative input

This example describes a task, not a tested output. Adjust the source material and settings to your own situation.

Prepare a private daily brief from these authorized sources, separating confirmed commitments from tentative ideas and taking no external action.

Result check: Check source freshness, remembered preferences and that no unapproved change or communication occurred.

Access and alternatives

The assistant has an MIT-licensed self-hosted codebase and paid managed plans. Free software does not include free infrastructure or model inference; hosted tiers also differ in compute, storage and usage. Confirm provider credentials, deployment maintenance and the currently supported client build before installing or subscribing.

Choose another workflow when its emphasis fits better: Letta for another stateful-agent approach; Lindy for a different connected-assistant workflow.

Illustrative workspace photo
Illustrative stock photo from Unsplash; not a screenshot of Vellum.

What you can do

Persistent assistant memory

Connected application tools

Scheduled and proactive tasks

Cross-device conversations

Hosted or self-hosted runtime

Where it helps

  • Memory supports continuity across repeated tasks
  • Self-hosting provides a distinct deployment choice
  • Connected tools can serve a clearly scoped routine

What to consider

  • Inference and infrastructure may cost separately
  • Remembered context and external actions require deliberate review

Pricing & free access

Official pricing
Free plan + paid

The assistant has an MIT-licensed self-hosted codebase and paid managed plans. Free software does not include free infrastructure or model inference; hosted tiers also differ in compute, storage and usage. Confirm provider credentials, deployment maintenance and the currently supported client build before installing or subscribing.

Plan information can change. Prices and conditions are set by the provider. Information checked 2026-10-15.

Source for access details

Get free-plan changes by email

One email a week: new free AI tools, prompts and free-plan changes. No spam, unsubscribe any time.

Questions about Vellum

Who is Vellum useful for?

Vellum's current offering is a personal assistant that can retain preferences and perform tasks across connected tools. It suits users who want continuity between conversations and repeated work. Start with a read-only task before enabling actions. Memory can retain an incorrect assumption as well as a useful fact, so inspect important context and set clear permission boundaries for messages, bookings and other outside changes.

Does Vellum have continuing free access?

The assistant has an MIT-licensed self-hosted codebase and paid managed plans. Free software does not include free infrastructure or model inference; hosted tiers also differ in compute, storage and usage. Confirm provider credentials, deployment maintenance and the currently supported client build before installing or subscribing.

How should I start a small evaluation?

Choose hosted access or review self-hosting requirements. Configure permitted sources and action boundaries.

What should I check in the result?

Check source freshness, remembered preferences and that no unapproved change or communication occurred.

When should I choose a different workflow?

Letta for another stateful-agent approach; Lindy for a different connected-assistant workflow.

Did this help you complete your task?

Tell us what needs improving. No account or private output required.

Was this useful?
Report incorrect information or a broken link

Review evidence

Documentation-based review · 2026-10-15

Reviewed from the linked official documentation. The workflow and example are editorial suggestions; no account-level product test or measured performance comparison was performed.

Put it to work

Other tools to consider

Explore all tools

Build stateful AI agents that retain memory across sessions, with a web app, CLI workflow, and developer API.

Best for: Letta suits builders who need an agent to maintain context beyond one conversation.

AI Agents & WorkflowsAI Infrastructure & Developer APIs

AI teammate for connected company tasks, including research, email, meetings, CRM updates, and computer-based work.

Best for: Connected app tasks and Research

AI Agents & WorkflowsProductivity & Automation

A free desktop app where several AI bots chat in one group, each using your own model and API key, with chats and files kept on your computer.

Best for: Teebots is a free local AI assistant for macOS, Windows and Linux.

AI Agents & WorkflowsProductivity & Automation

We use optional cookies for analytics and referral tracking. Learn more about our privacy policy