Skip to content

Ainvented pricing

Pay per million tokens, by model tier: input, cached input and output. There is no monthly plan; project budgets and hard caps keep every agent's spend visible. Videos, images and voice notes are billed as the input tokens they become. See the model tiers and the services.

Straightforward pricing

Pay for what agents use. Nothing they don't.

Choose between four Ainvented models. Input, cached-input, and output rates refresh from API Hub while project budgets keep spend visible.

  • Visual agent builder and production API
  • Project budgets and hard caps
  • Token, cost, and usage reporting
  • Cached-input savings and auto-recharge
Build your first agent
Four public models

Choose how each agent runs.

Live from API Hub

Epic

Highest-capability model

ainvented-epic
Input
$4.00/ 1M tokens
Cached input
$0.80/ 1M tokens
Output
$20.00/ 1M tokens

Saga

Advanced reasoning model

ainvented-saga
Input
$2.00/ 1M tokens
Cached input
$0.40/ 1M tokens
Output
$6.00/ 1M tokens

Lite (Default)

General-purpose model

ainvented-default
Input
$0.30/ 1M tokens
Cached input
$0.060/ 1M tokens
Output
$1.50/ 1M tokens

Low

Lower-cost model

ainvented-low
Input
$0.20/ 1M tokens
Cached input
$0.045/ 1M tokens
Output
$0.65/ 1M tokens

Cached input is the reduced rate for eligible reused context.

USD · No required monthly plan