Additional tokens can only be purchased by pro or enterprise plans.
If you are on a paid plans follow the instructions below to purchase more tokens. The price for additional tokens is $5 per 1 Million tokens.
Pro & Business
Hover on the AI icon in your Assistant's sidebar and click on the 'Purchase Tokens' button.
Choose the number of tokens you'd like to purchase and hit confirm purchase.
Once you've confirmed your billing details, complete your purchase.
You'll see your new token balance reflected shortly! (May take up to 1 minute)
Note: Only the admin of the workspace may purchase tokens.
If you are on an enterprise account, please contact your account representative that can facilitate the purchase and add additional tokens to your account.
Free Token Quota
Each plan has a different free token quota. This quota represents the free monthly tokens that are included with your account. These reset each month.
- Starter Plan: 100,000 Tokens
- Pro Plan: 1,000,000 Tokens
- Business: 5,000,000 Tokens
- Enterprise: Custom
Additional tokens purchased are in addition to your monthly quota. They are a one time purchase and do not expire or reset. You will see them added to your monthly token quota as in the example below.
- On a pro plan, I have used 500,000 of my 1,000,000 monthly quota on July 14. My token meter current shows [500,000/1,000,000 used].
- If I purchase an additional 1,000,000 tokens on July 14th, my token meter will show [500,000/2,000,000 used].
- If I use 1 million additional tokens, my token meter will show [1,500,000/2,000,000 used]. This means I have used 1 million of the free tokens and 500k of the purchased tokens.
- On August 1st my quota will reset and it will display the free amount + what remains of my purchased amount. In the example above, my token meter would show [0/1,500,000 used]. This is the 1 million free tokens I get each month + the 500k remaining from my purchased amount.
How are tokens calculated?
What is a token?
A token is determined by the AI model you use. For example, OpenAI (GPT models) and Anthropic (Claude models) considered a token as around 3-4 characters. To read more on how tokens are calculated you can read the documentation from either AI provider below.
More powerful AI models use more tokens
In Voiceflow, you can use different AI models to answer your question. These different AI models have vastly different costs.
In Voiceflow this cost is represented by multipliers. More accurate models use more tokens.
- GPT-3.5: 1x
- Claude Instant: 1x
- Claude V1: 10x
- GPT-4: 25x
For example, using GPT-4 will use 25x more Voiceflow AI tokens than GPT-3.5 for the same question.
Keep this in mind when you are designing your flow. Only use GPT 4 when you require high accuracy and be mindful of how many tokens you use. Otherwise you might use up your tokens quickly!
How are tokens calculated?
Tokens are used by both the input (what is being sent to the AI model) and output (what is being received to the AI model).
It is important to note, that the 'token' slider on the AI steps (shown below) only control the output message. Depending on your design, there may be a large amount of information that is sent to the AI model as a part of your prompt
How many tokens do different features use in Voiceflow?
- Knowledge Base: A question and answer will use anywhere from 800-1500 tokens depending on the size of question and answer. This is because the Knowledge base passes in relevant information from your documentation to the AI model alongside your prompt.
- Response AI Step (With Memory): A question and answer when using the Response AI step with Memory will use anywhere from 150-2000 tokens. Voiceflow saves the last 3 conversation turns and passes those to your AI model to provide context for your answer.
- Response AI Step (Without Memory): A question and answer when using the Response AI step without, memory will use roughly 150-300 tokens.
You can turn memory on or off via the setting on the Set AI and Response AI step below.
All Token limits are subject to change on General Release as AI Assistants are currently in Beta.