Free for small teams. $10/user for full history and production agents.
Free includes up to 10 groups and unlimited users. Agent model costs are billed directly and visible in Settings.
Seat price covers the workspace. Agent costs are separate and scale with what the agent actually does — comparable to running that same model directly, whether it is summarizing one thread or drafting updates across a whole workspace. Agents can use Anthropic’s Haiku, Sonnet, or Opus, or OpenAI’s GPT-5.6 tiers — Sol, Terra, and Luna — chosen automatically based on the work. We bill you the model costs directly, at the same rate the provider charges us. Free includes test-only credits so you can try Loop Agent and MCP without production spend. On Standard and Enterprise, production agent usage is billed as you go and visible in settings, and you can set spending limits at any time; contact us if you want to discuss volume pricing.
Illustrative examples at current model rates — your actual bill depends on message length and how often agents run.
Set a hard spending limit for the whole workspace or per agent in Settings → Billing at any time. Bring-your-own provider key is on our roadmap; today all usage is billed directly by FabricLoop at provider rates.