Units

Every query to the AI incurs a cost in Units. Upon creating an account, we provide some free units. Subsequently, a subscription plan is needed to receive a monthly allocation of units, varying based on your chosen plan.

For Pro plan users, units are not required if you set up your own Open AI API Key.

Tokens

It's important to distinguish between a unit and a token.

A token refers to a basic unit of text, typically used to measure the input and output of an AI model. Here's a more detailed explanation:

• Basic Unit of Text: A token is usually a word, but it can also be a punctuation mark or a symbol. In some cases, particularly in languages where words are not separated by spaces, a token might be smaller than a word, like a syllable or even a single character.
• Measuring AI Interactions: In AI models, especially those dealing with language (like GPT-3 or GPT-4), tokens are used to quantify the amount of text processed. For instance, when you input a sentence into the model, it is broken down into tokens, and each token is analyzed.

While the number of tokens used reflects your consumption, it does not directly indicate the cost of what you've used.

Units vs Tokens

Under the hood, the calculation of unit consumption is based on token consumption. The value of a unit is linked to the real price of a token in a specific context. In essence, a unit represents the token's value. For instance, a token utilized with the latest and most advanced AI model will be more expensive than one used with an older model.

Cost in Units

Here's a table summarizing the cost in units depending on the AI model used:

You have the ability to manage your unit costs through various mechanisms:

A max token setting is available when executing an Instruction, designed to halt the query should it surpass a predefined resource limit. Within the Flow Builder, it's possible to set resource limits for each Question or for the entire Flow. We strongly advise prudent use of these settings to prevent unexpected outcomes and as a safeguard. It's important to note that these settings are based on tokens, not units.

Reserved Units

Each time you execute a query, you define a maximum token limit that the query can use. Based on this limit,we reserve the maximum units that could potentially be consumed by the query. Once the query concludes (whether successful or not), we calculate its actual cost. This cost is then billed, and any excess reserved units are returned to you.

Please be aware that sometimes units might temporarily remain in reservation mode. Typically, they are released once the transaction associated with your queries is resolved, which may take a few minutes, especially if there are issues.

Pro Plan

The Pro Plan offers the option to use your own API key or rely on units for queries. If you opt for your own API key, Open AI will bill you directly based on usage. Without your own API key, units will still be available for query executions.

You have the flexibility to switch between units or tokens consumption mode under the Pro Plan. Using specific services such as Image Generation or Newsroom Nodes requires consumption of units. Therefore, even with a Pro Plan, units are needed to access all the exciting features.

Receiving Units

Acquiring units necessitates a subscription to one of our plans. Upon subscribing, you're immediately allocated units corresponding to the chosen plan. Subsequently, the same number of units will be credited to your account at the start of each billing cycle (monthly), continuing until the subscription is canceled.

It's important to note that updating your plan mid-cycle doesn't accrue units from the previous plan. Effectively, your initial plan is terminated, and you embark on the new plan immediately. This change resets your billing cycle, which begins anew from the first payment date of the new subscription.