Every model, one key.

Or let Flux pick

Send one of these names instead of a model id. Flux picks among vetted models for each kind of task, cost-aware, and bills the lane's flat rate.

• AutopilotRecommended

  flux-auto

  Flux reads each request and picks a model suited to the job. The recommended choice.

  $2.00 in, $8.00 out, per 1M tokens

• Reasoning

  flux-reasoning

  For hard problems: planning, debugging and long chains of logic.

  $4.00 in, $15.00 out, per 1M tokens

• Standard

  flux-standard

  Everyday work, balanced between quality and cost.

  $2.00 in, $8.00 out, per 1M tokens

• Fast

  flux-fast

  Quick, low-cost calls such as short edits, lookups and summaries.

  $1.00 in, $4.00 out, per 1M tokens

Autopilot (flux-auto) bills at the Standard rate, or the Fast rate when it sends your request to the Fast lane and a Fast model answers. Never more than Standard. Pay as you go adds 12% to these rates. Pinned models bill at their own rate. See pricing.

Pin a model by lab

Put the id in your agent's model setting. Your requests go to that model and no other.

The catalogue also holds 24 image and 6 voice models. They sit behind the image and voice abilities, not the chat model setting.