Questions, answered.
The things people ask us most about Chorus. If you don't see yours here, write to support@chorus.com.
The basics
Chorus is an agent platform. Each agent runs on its own dedicated VM with a real browser, 70+ integrations, and typed memory — so it can ship code, run research, design assets, and operate tools end-to-end.
ChatGPT and Claude are chat products. Chorus is closer to giving an employee their own computer: your agent can keep files and memory, use its own email or connect to yours, and work across the tools your business already runs on — Linear, Notion, Docs, Excel, QuickBooks, and more.
No. Most agents are configured in plain English. If you want to go deeper, you can install skills, wire up your own integrations, or hand-write prompts — but you don't have to.
Under a minute to your first running agent. Pick a template, sign in, and you're talking to it.
Money stuff
Chorus has two paid plans: Pro is $20/month for 1 always-on agent, and Max is $200/month for 3 always-on agents. Both include recurring usage credits; extra agents are $20/month each. See the pricing page for the full breakdown.
Yes. The Free plan includes 1 agent, $5 in monthly usage credits, and no card required. It is meant for trying the workflow before upgrading to Pro or Max.
Paid plans renew monthly. Your included monthly credits reset on your billing date, and Pro and Max also receive additional credits every 5 hours for active work. You can manage your plan, credits, and extra agents from billing.
Your agents keep their memory, files, and setup, but model and media work can pause until you add credits or enable auto-reload. Once credits are available again, they can pick up where they left off.
What an agent can do
Anthropic's Claude Opus 4.8 and Sonnet 4.6, OpenAI's GPT-5.5, and Google's Gemini 3.1 Pro for text. For images, Google's Nanobanana Pro and OpenAI's GPT-image-2. Pick a default per agent or let Chorus route automatically based on the task. Full pricing on the pricing page.
70+ out of the box — GitHub, Linear, Slack, Notion, Google Workspace, Figma, Stripe, and the long tail. Each agent picks the ones it needs and Chorus handles the auth.
Yes. Every agent has a real browser inside its VM. It can log into sites, fill forms, click buttons, take screenshots — anything you can do in a tab.
Yes, and they can talk to each other. The group-chat surface is built so one agent can hand off work to another mid-conversation.
Where your data lives
Each agent runs in its own isolated VM on our infrastructure. Memory and files persist on encrypted volumes scoped to that agent. Nothing is shared across accounts.
No. We don't train models on your data, and the model providers we route to are configured for zero-retention where they offer it.
We're actively in the audit window. Reach out if you need the current report or our security questionnaire — see the security & platform page for details.
Yes. Delete an agent and its VM, memory, and files are wiped. Delete your account and the whole footprint goes with it.