Every Developer seat comes with $10 of usage per month. Once that's used up, anything you run on the Codistry key is charged in dollars, and the cost of a request covers two things:

• Model cost — whatever the provider charges for the tokens, at their published rates
• Codistry fee — 25% on top of that model cost

If you bring your own key or run local models, there is no fee at all.

Automatic top-ups are on by default, so a small amount is added when your balance runs low — you can change the amounts or turn them off from your dashboard, and add credits manually whenever you want more. You can also switch any session to your own key or a local model and keep working without credits.

Any of them. Use hosted models on the Developer plan, bring your own Anthropic or OpenAI key, route through OpenRouter, or point Codistry at any OpenAI-compatible endpoint, including local models through Ollama or LM Studio. You can switch models mid-session, per prompt.

Everything Codistry learns about your code lives in your editor and your repo. Cloud models see only what you choose to send, and with local models it runs fully offline, even air-gapped. The telemetry docs list exactly what the extension sends.

Yes. You can cancel or downgrade from your dashboard at any time, and your plan stays active through the period you already paid for.

These are founding prices and packages for our first set of customers. They're set for a limited time and will be updated later as the product evolves.

Yes. As part of the limited-time launch pricing you can bring your own model into Codistry for free, local or cloud.

Free and Developer are individual plans. For teams up to 100 users, Teams & Enterprise adds user management, team-level custom prompts, volume token discounts, on-prem or air-gapped deployment, and SSO/SAML with SCIM and audit export. Contact sales and we'll set it up together.