Simple, honest pricing

Credits are Vellum's billing unit. $1 = 1 credit. They are consumed by LLM inference, web search, image generation, and any paid third-party APIs routed through managed OAuth. Credits are pay-as-you-go: you only pay for what your assistant actually uses. See pricing docs.

Machine size sets the compute ceiling for your assistant: the maximum CPU and RAM available at runtime. Base includes Small (fixed). Pro unlocks Medium (2.5 vCPU / 5 GiB RAM), Large (4 vCPU / 8 GiB RAM), and XL (4 vCPU / 16 GiB RAM). You can resize anytime from your assistant's settings page.

The $10/mo base fee covers your assistant's custom subdomain, a static IP address, and priority support. It applies to all Pro subscriptions on top of any machine and storage add-ons.

Yes. All plan changes live in Settings > Billing. Upgrades are active immediately. Machine and storage tier changes are prorated. Downgrades or cancellations take effect at the end of your current billing period.

Persistent disk: files, notes, conversation history, skills, and any data your assistant writes to the workspace volume. Storage grows online, no assistant restart needed.