Skip to content
SPECIFICTION
Get listed

Vellum

vellum.ai

From $30/mo Free plan Open source WebWindowsmacOSiOSAndroidAPI
Visit vellum.ai
Spec score 8.6 #2 of 49 in LLM Application Development Frameworks

What it is

Vellum is a personal AI assistant that can read files, manage calendars, send email, run code, and control a computer. It saves preferences and context in its workspace for later conversations, and users can create, install, and manage skills. Listed OAuth services include Discord, GitHub, Google services, HubSpot, Linear, Twitter (X), Asana, Notion, Microsoft Outlook, and Todoist. Vellum says OAuth tokens stay in a credential vault rather than being exposed to the language model. Sensitive actions prompt for approval, and users can set trust rules for standing permissions. By default, conversation context is sent to Anthropic to generate responses; OpenAI, Google, OpenRouter, Fireworks, and local Ollama are listed alternatives. An internet connection is required. Vellum Cloud runs assistants in isolated environments with private storage and 24/7 availability. The Free plan includes first tasks and pay-as-you-go credits. Mighty costs $30 a month, Super $100 a month, and Ultra $200 a month. There is no shipped Linux desktop client, but Linux users can use the web app, Chrome extension, or self-host the runtime.

Spec sheet

What it does
Vellum is a personal AI assistant that can read files, manage calendars, control computers, take calls, perform transactions, and retain personal context.Read 15 Oct 2026 · vellum.ai
Product
Vellum describes itself as a personal AI assistant that can read files, manage calendars, send email, run code, and control a computer.Read 30 Sep 2026 · vellum.ai
Audience
Vellum characterizes the product as a personal assistant rather than a team or shared company tool.Read 30 Sep 2026 · vellum.ai
Integrations
OAuth integrations include GitHub, Notion, Linear, Airtable, Spotify, Gmail, Google Calendar, Slack, and other services, while MCP servers provide additional tools.Read 15 Oct 2026 · vellum.ai
Support
Vellum lists support contact as [email protected].Read 30 Sep 2026 · vellum.ai
Headquarters
United StatesRead 19 Oct 2026 · vellum.ai
AI data routing
By default, conversation context is sent to Anthropic to generate responses; the docs also list OpenAI, Google, OpenRouter, Fireworks, and local Ollama as alternatives.Read 30 Sep 2026 · vellum.ai
API maturity
The API documentation marks endpoints as Beta or generally available and says the surface is still expanding as APIs stabilize.Read 15 Oct 2026 · docs.vellum.ai
Approval controls
Sensitive actions prompt the user for approval, and users can configure trust rules for standing permissions.Read 30 Sep 2026 · vellum.ai
Built-in capabilities
Documented capabilities include Gmail, Google Calendar, Slack, web browsing, coding, app building, image generation, transcription, scheduling, screen watching, and media processing.Read 15 Oct 2026 · vellum.ai
Cloud hosting
The default deployment provisions an always-on assistant in an encrypted, account-isolated Vellum Cloud workspace with no server administration required.Read 15 Oct 2026 · vellum.ai
Communication channels
The assistant is reachable through the web, desktop and mobile apps, Telegram, Slack, voice input, and phone calls through Twilio.Read 15 Oct 2026 · vellum.ai
Company funding record
Crunchbase lists Vellum as funded by 11 investors across three rounds, with Leaders Fund and Y Combinator among the latest investors.Read 15 Oct 2026 · crunchbase.com
Company legal name
Vellum's privacy notice identifies Vocify, Inc. as the provider of the service.Read 15 Oct 2026 · vellum.ai
Company profile
LinkedIn lists Vellum as privately held, founded in 2023, headquartered in New York, and employing 11–50 people.Read 15 Oct 2026 · linkedin.com
Company security commitments
On its About page, Vellum says it encrypts data end-to-end, conducts continuous security audits, and does not train on personal information.Read 30 Sep 2026 · vellum.ai
Computer control
Desktop apps can click, type, and navigate supported Mac or Windows applications, with sensitive actions gated by approval.Read 15 Oct 2026 · vellum.ai
Credential protection
Credentials are kept in an isolated credential service and are not exposed to the assistant or model; self-hosted credentials use the macOS Keychain or encrypted storage.Read 15 Oct 2026 · vellum.ai
Custom skills
Users can create, install, and manage skills to extend the assistant's capabilities.Read 30 Sep 2026 · vellum.ai
Documentation depth
Developer documentation covers architecture, security, integrations, skills, browser automation, attachments, event streaming, deployment, SDKs, and API references.Read 15 Oct 2026 · vellum.ai
Extensibility
Users can create custom skills and connect MCP servers when an existing built-in tool does not cover a task.Read 15 Oct 2026 · vellum.ai
Funding
TechCrunch reported a $5 million seed round announced July 11, 2023, with participation from Y Combinator, Rebel Fund, Pioneer Fund, Eastlink Capital, and angels.Read 15 Oct 2026 · techcrunch.com
Gartner review status
Gartner Peer Insights lists Vellum but reports no reviews yet.Read 15 Oct 2026 · gartner.com
Google Play data disclosure
Google Play says the Android app may share location, personal information, and app activity, may collect personal information and additional data types, encrypts data in transit, and supports deletion requests.Read 15 Oct 2026 · play.google.com
Integration security
Vellum says OAuth tokens are stored in a credential vault and are not exposed to the language model.Read 30 Sep 2026 · vellum.ai
Licence
The Vellum Assistant repository is open source under the MIT licence; the managed Vellum Platform is a separate commercial offering.Read 15 Oct 2026 · github.com
Limit: Apple sign-up and language
Apple reviews report an earlier sign-up failure and complain that the interface initially lacked Chinese-language support.Read 15 Oct 2026 · apps.apple.com
Limit: evaluation tooling
G2 reviewers say the evaluation suite and evaluation UI lag specialized competitors and would benefit from better annotation, datasets, and feedback workflows.Read 15 Oct 2026 · g2.com
Limit: interface polish
A G2 reviewer reports that the interface can be clunky and buggy, while another says workflow UX is not yet intuitive.Read 15 Oct 2026 · g2.com
Limit: learning curve
A Capterra reviewer identifies a learning curve for Vellum's more advanced features.Read 15 Oct 2026 · capterra.com
Linux availability
There is no shipped Linux desktop client; Linux users can use the web app, Chrome extension, or self-host the runtime.Read 30 Sep 2026 · vellum.ai
Linux desktop limitation
There is no shipped Linux desktop client, although Linux users can use the web app, Chrome extension, or self-hosted runtime.Read 15 Oct 2026 · vellum.ai
Memory
The assistant saves preferences and context to its workspace for use across conversations.Read 30 Sep 2026 · vellum.ai
Model data handling
Conversation context is sent to Anthropic for inference, while Vellum says it does not use customer data for training and telemetry is opt-in.Read 15 Oct 2026 · vellum.ai
Notable news
On April 21, 2026, Vellum released one-click Chrome Web Store installation, generally available X integration, expanded Slack permissions, and web voice settings.Read 15 Oct 2026 · vellum.ai
Official SDKs
Vellum publishes official client libraries for Python, Node/TypeScript, and Go.Read 15 Oct 2026 · docs.vellum.ai
Offline use
Vellum says the assistant cannot respond without an internet connection because it needs to reach an AI model provider.Read 30 Sep 2026 · vellum.ai
Persistent memory
The assistant stores preferences, projects, people, and other context across conversations and retrieves memories when relevant.Read 15 Oct 2026 · vellum.ai
Praise: Apple reviewers
Apple reviewers praise the polished interface, personalization, broad capability, and ability to act through APIs, MCP servers, and skills.Read 15 Oct 2026 · apps.apple.com
Praise: Capterra reviewers
Capterra reviewers praise Vellum's model coverage, observability, client SDK, drag-and-drop workflows, and rapid pipeline iteration.Read 15 Oct 2026 · capterra.com
Praise: support
G2 reviewers repeatedly describe Vellum staff as responsive, helpful, and quick to address implementation questions and product feedback.Read 15 Oct 2026 · g2.com
Praise: workflow speed
G2 reviewers repeatedly praise the visual workflow builder, rapid prompt and model comparison, one-click deployment, and faster movement from experimentation to production.Read 15 Oct 2026 · g2.com
Public API
The documented API uses versioned HTTP endpoints, with cloud requests normally sent to api.vellum.ai and separate base URLs for self-hosted deployments.Read 15 Oct 2026 · docs.vellum.ai
Rating on Apple App Store
4.4 out of 5 from 17 ratingsRead 15 Oct 2026 · apps.apple.com
Rating on Capterra
4.8 out of 5 from 8 reviewsRead 15 Oct 2026 · capterra.com
Rating on Chrome Web Store
5.00 out of 5 from 3 ratingsRead 15 Oct 2026 · chrome-stats.com
Security approvals
Sensitive actions such as sending mail, purchases, and desktop control require explicit approval and can be governed by trust rules.Read 15 Oct 2026 · vellum.ai
Self-hosting
The runtime can be self-hosted locally or on user infrastructure, including Linux hosts and customer VPC deployments.Read 15 Oct 2026 · vellum.ai
System requirements
The downloads page requires iOS 15 or later, Android 7 or later, macOS 15 Sequoia or later, Chrome 120 or later, and Windows 10 or later.Read 15 Oct 2026 · vellum.ai
Training and telemetry
Vellum says it does not use user data for model training, and optional usage analytics and crash diagnostics are off by default.Read 30 Sep 2026 · vellum.ai

Plans and prices

Base Free Free, forever 1 vCPU · 2 GiB RAM · 6 GiB persistent storage · pay-as-you-go credits · managed LLM credentials Read 15 Oct 2026 · vellum.ai
Free Free Start free; first tasks included; pay-as-you-go credits Read 30 Sep 2026 · vellum.ai
Mighty $30 a month $30/mo Small machine · 10 GiB storage · Mighty Usage · platform fee not included · no custom subdomain, static IP, or priority support Read 15 Oct 2026 · vellum.ai
Mighty $30 a month $30/month Small computer 1 vCPU / 2 GiB · 10 GB storage · Mighty Usage included Read 30 Sep 2026 · vellum.ai
Custom $50 a month Minimum $50/mo; credits pay-as-you-go Choose machine, storage, and optional usage bundle · platform fee $10/mo · machines from Medium to XL · storage 10 to 120 GiB Read 15 Oct 2026 · vellum.ai
Super $100 a month $100/month Medium computer 2.5 vCPU / 5 GiB · 30 GB storage · Super Usage included · Assistant email and subdomain Read 30 Sep 2026 · vellum.ai
Super $100 a month $100/mo Medium machine · 30 GiB storage · Super Usage · platform fee included · custom subdomain · static IP · priority support Read 15 Oct 2026 · vellum.ai
Ultra $200 a month $200/month Large computer 4 vCPU / 8 GiB · 60 GB storage · Ultra Usage included · Assistant email and subdomain Read 30 Sep 2026 · vellum.ai
Ultra $200 a month $200/mo Large machine · 60 GiB storage · Ultra Usage · platform fee included · custom subdomain · static IP · priority support Read 15 Oct 2026 · vellum.ai

Every Vellum plan and price →

Strong points and limits

Strong points

  • Saves preferences and context across conversations.
  • Users can add and manage custom skills.
  • Sensitive actions prompt for approval.
  • OAuth tokens are stored in a credential vault.
  • Optional analytics and crash diagnostics are off by default.

Limits

  • The assistant requires an internet connection.
  • There is no shipped Linux desktop client.
  • Conversation context goes to Anthropic by default.
  • Paid plans start at $30 a month.

Who it suits

Vellum suits an individual looking for an assistant that can handle files, calendars, email, code, and computer control. The maker characterizes it as a personal assistant rather than a shared company tool.

Vellum in depth

Read the full write-up

Overview

Vellum is a personal AI assistant for people who want software to take action across everyday digital tasks, not just answer questions. It is best suited to individuals comfortable configuring an assistant and weighing what to let it do.

Its combination of persistent context, custom skills, and approval controls gives it a wider role than a conventional chat assistant, while its internet dependency and personal-use focus make it a poor fit for offline work or shared company workflows.

Key features

Vellum can read files, manage calendars, send email, run code, and control a computer. Its workspace retains preferences and context between conversations, which can reduce repeated setup for recurring tasks. Users can also create, install, and manage skills; that flexibility is useful for tailoring the assistant, but depends on users being willing to shape its capabilities.

OAuth connections cover Discord, GitHub, Google services, HubSpot, Linear, Twitter (X), Asana, Notion, Microsoft Outlook, and Todoist. Vellum says their tokens remain in a credential vault rather than being exposed to the language model. Sensitive actions require approval, while configurable trust rules allow standing permissions. This is a sensible balance for software that can act on a user's behalf, though users still need to judge which actions merit ongoing trust.

Vellum Cloud runs assistants in isolated environments with private storage and 24/7 availability. Conversation context goes to Anthropic by default; OpenAI, Google, OpenRouter, Fireworks, and local Ollama are other provider options. The choice of provider does not remove the internet requirement: Vellum says it cannot respond without a connection to an AI model provider. The company says it does not use user data for model training, encrypts data end-to-end, and conducts continuous security audits. Optional usage analytics and crash diagnostics are off by default.

Pricing

The free plan costs nothing and includes first tasks, with additional use through pay-as-you-go credits. It is a way to start without a subscription, but the included tasks and credit arrangement make it less predictable for sustained use.

Mighty costs $30 a month, billed $30/month. It includes a small computer with 1 vCPU and 2 GiB, 10 GB storage, and Mighty Usage. This is the entry paid tier for an individual who needs a continuously available assistant with modest compute resources.

Super costs $100 a month, billed $100/month. It raises the allocation to a medium computer with 2.5 vCPU and 5 GiB, 30 GB storage, and Super Usage; assistant email and a subdomain are included. It suits users who need more hosted capacity and those additional assistant-facing features.

Ultra costs $200 a month, billed $200/month. It provides a large computer with 4 vCPU and 8 GiB, 60 GB storage, Ultra Usage, assistant email, and a subdomain. This tier is aimed at heavier hosted workloads; the higher price is difficult to justify if the smaller computer is sufficient.

The plan descriptions do not give numerical usage quotas, so buyers should not assume a fixed task allowance from the included Usage labels. Support contact is [email protected].

Platforms

Vellum lists Android, iOS, macOS, Windows, web, extension, and self-hosted availability. There is no shipped Linux desktop client; Linux users can use the web app, Chrome extension, or self-host the runtime. The assistant's internet requirement also rules it out for offline operation.

Who it's for

Vellum is aimed at individuals who want a configurable assistant to work across connected services and retain useful context, and who value approval prompts before sensitive actions. It is less suitable for teams seeking a shared company tool, people who need offline responses, or Linux users who require a native desktop client.

Pros and cons

Pros: Persistent context and customizable skills support recurring, personalized tasks rather than one-off chat alone.

Pros: Approval prompts, configurable trust rules, and credential-vault handling address the risks of granting an assistant access to connected services.

Pros: Cloud hosting offers isolated environments, private storage, and 24/7 availability.

Cons: Responses require an internet connection to reach an AI provider, so Vellum cannot serve as an offline assistant.

Cons: The paid plans start at $30 a month and scale to $200 a month, while their Usage allowances have no numerical quotas in the plan descriptions.

Cons: Linux has no native desktop client, and the product is positioned for personal rather than shared team use.

Alternatives

For prompt and agent tooling, Promptitude is worth considering when a reader needs a web or API service with a free tier capped at two users, five prompts, two assistants, one flow, one tool, 100 generations per month, and 200 tool calls per month. Agenta is a closer option for team-oriented agent work: its free Hobby plan includes two team members, 5,000 agent runs per month, 20 evaluations per month, and one-week trace retention, with API, self-hosted, and web platforms.

For a free, self-hostable prompt-learning framework rather than a personal assistant, OpenPrompt runs on Linux, macOS, Windows, or self-hosted environments and specifies Python 3.8+ and PyTorch 1.8.1+ as tested requirements. PromptWizard is another free option with API, Linux, macOS, Windows, and self-hosted platforms. Readers comparing other prompt, evaluation, or agent tools can browse AI Agent Platforms, AI LLM Evaluation Tools, AI Prompt Generators, AI Prompt Management Software, AI Workflow Automation Software, and Prompt Management Software.

Verdict

Choose Vellum if you are an individual seeking an assistant that can work across files and connected services, preserve context, and pause for approval before sensitive actions. Its main trade-off is the combination of provider-dependent internet access and paid hosted tiers starting at $30 a month; look elsewhere if offline use, team collaboration, or a Linux desktop client is essential.

Verdict

Vellum combines a broad set of assistant tasks with persistent context, skills, and approval controls. Consider its internet requirement and default AI provider routing, and note that Linux access uses the web app, extension, or self-hosted runtime rather than a desktop client.

Where Specifiction places it

Alternatives to VellumAll alternatives →