Dropstone Support

Dropstone as an alternative to Claude Code

If you are weighing Dropstone against Claude Code, the decision usually turns on whether the agent remembers across the tools you use, whether one account covers chat and the terminal, and what the organization controls look like. Here is our side of those, including where Dropstone is the wrong choice.

People arrive here having heard of Claude Code and wondering whether Dropstone fits better. The question usually turns on three things: whether the agent remembers across the tools you use, whether one account covers chat and the terminal, and what an administrator can control. This page answers those on our side, plainly, and says where Dropstone is the wrong choice.


What is different

One memory, every surface. One memory. Every surface. Teach it a rule in the terminal and the chat app already knows it. Correct it in chat and your own code, running through the SDK, follows it. Nothing has to be re-explained per tool. See Memory across chat, CLI, and SDK.

One account, not one tool. The same account covers Dropstone Chat, the CLI, the SDK, and the editor extension. Your plan, your usage allowance, and your memory are the same ones on each.

Models that are chosen for you, and re-chosen monthly. Three tiers, Fast, Pro, and Heavy. Every month the strongest model available for each tier is re-evaluated and becomes that month's pick, so the name stays the same while what sits behind it improves. See Choose a model.

An organization console, not just seats. Members and teams, roles, policies for models and connectors, the audit log, single sign-on, SCIM provisioning, and content logging for compliance. See Dropstone for teams and organizations.

Where it runs. Requests are processed in the United States on certified providers with zero retention upstream, and enterprise and commercial data is excluded from training. See Where your requests run.


What Dropstone gives you, on the axes people compare

QuestionDropstone
Which surfacesChat, CLI, SDK, editor extension, one account
Does it rememberYes, one memory per account, read and written on every signed-in surface
Which modelsFast, Pro, and Heavy, re-evaluated every month
For organizationsRoles, policies, audit log, SSO, SCIM, content logging
From your own codeA typed SDK with the same memory; the headless API is stateless by design
Where it runsUnited States, zero retention upstream
What it costsdropstone.io/pricing

Where Dropstone is the wrong choice

  • You want to bring your own model or provider. Dropstone serves its own three tiers and you pick a tier, not a provider. If your requirement is to point the agent at a model of your choosing, this is not that.
  • You want nothing remembered anywhere. A stateless completions endpoint is a smaller thing than what Dropstone is for. The API is stateless, but every signed-in surface is not.

How to start

  1. Install the CLI and run dropstone login. It opens your browser once and returns to the terminal.
  2. Ask it something real from your own project, then correct it once when it gets something wrong. That correction is what it carries from then on.
  3. Open chat.dropstone.io and ask what it remembers. It should tell you what you just taught it, which is the whole idea.

Related articles

Ctrl+I