How does reasoning appear in a stream?

On Max and Heavy, chunks first carry delta.reasoning_content (the model's thinking), then delta.content (the answer). Flash doesn't think at length, so it goes straight to content.

• Show reasoning in a collapsible "thinking" area, or ignore it.
• It's billed as output tokens either way.
• If you only print delta.content, the first seconds on Heavy may look silent while it thinks. That's normal.

Good to know#

• Your key's account is responsible for your end users' text under the Abuse policy. What it means for apps
• Model ids: supercosmos-flash, supercosmos-max, supercosmos-heavy (SuperCosmos 1.1). Which to call
• Keep keys on your server, never in a browser or mobile app. Why

Still need help?#

• General support: support@clatterlabs.com or team@clatterlabs.xyz
• Infrastructure questions, or permanent deletion of your data (your whole account or one service): infrastructure@clatterlabs.xyz

We usually reply within the first hour, and always within 24 hours. A full support team is with you.