In Hatz, a model's multiplier tells you how much more or less a model will use per message compared to the industry's flagship model, which sits at 1x. A model at 0.5x uses about half the credits per message; a model at 2x uses about double. Use the multiplier to compare models at a glance before you pick one.

You'll see multipliers in two places in the model selector:

The multiplier is a reliable estimate of how much more or less a model will cost per message relative to the flagship. Models below 1x use fewer credits per message than the flagship. Models at 1x and above match the flagship or use more.

Your actual credits used on any request still depend on the tokens you send and receive at the model's per-token rate, so two requests on the same model can spend different amounts. The multiplier gives you a faithful comparison before you choose, without having to do token math.

Hatz calculates each multiplier from the current credit rates for that model and displays it as a comparison against the reference model. The multiplier is meant to help you compare models directionally, not predict an exact credit amount for a specific request.

If usage is higher than expected, compare the models being used in chat, workflows, apps, agents, and API requests. Auto Mode can help route requests to an appropriate model for the task. Admins can also use org or role Auto Mode policies and disable higher-multiplier models that a tenant does not need. For high-volume work, also review prompt length, files, tools, retries, and workflow step count.

From time to time, Hatz runs promotional pricing on specific models, temporarily lowering their per-token credit rate. The displayed multiplier always reflects the current rate, so a model on promotion will show a lower number for the duration of the promo. The model selector and Multipliers table also flag promotional models with a Promo badge, so you can see at a glance which ones are discounted.