Censored preview
cortex-1.2
$0.50 / 1M input · $1.50 / 1M output · $0.05 / 1M cached
1M context. 256k output. Same crew. You can switch models in a chat at any time.
Docs
Models
Two models. cortex-1.2 is a censored preview. cortex-1.1 is stable. You pay per token.
Stable
cortex-1.1
$0.50 / 1M input · $1.50 / 1M output · $0.25 / 1M cached
1M context. 256k output. Same crew. Coordinator and Coder on your files. No cold start.
How you pick
In the app the dock lists Cortex 1.2 (CENSORED PREVIEW) and Cortex 1.1. You can switch in a chat at any time. On the API, send puck-cortex-1.2 or cortex-1.2 for the censored preview, puck-cortex-1.1 or cortex-1.1 for stable. puck-core and puck-max still run cortex-1.1.
What you pay
No subscription. Each call is billed at the rate of the model you picked, rounded to the nearest $0.0001. Files stay on your computer. Only prompts leave the machine. Nothing is kept after.
- cortex-1.2Censored preview · $0.50 / 1M input · $1.50 / 1M output · $0.05 / 1M cached · 1M context · 256k output
- cortex-1.1Stable · $0.50 / 1M input · $1.50 / 1M output · $0.25 / 1M cached · 1M context · 256k output
- API idspuck-cortex-1.2 · puck-cortex-1.1
- UncensoredNo refusal filters.