Costs per user will vary depending on usage. You can use the calculator provided on Passio's account portal to estimate token usage.

‍
For example, if a user sends one photo to the API for analysis, it will result in 5k tokens spent from the account. So while costs can vary, we estimate that an average user logging meals regularly for one month would utilize around 250k tokens.

‍
To manage costs, you can limit usage to a fixed amount by engineering your app with token limits. For instance, you could limit a free user to 80k - 100k tokens and then ask users to pay once they reach that limit. This approach allows you to demonstrate value to users while keeping your cost between 1 to 10 cents per user.

Tokens are used to quantify the usage of Passio services. Each action, such as an API request, Vision AI request, Text AI request, or Chat Message request, consumes a specific number of tokens. The monthly plans include a set number of tokens, and users can purchase additional tokens if they exceed their monthly limit.

A token is a unit of measurement used in AI processing to represent pieces of input or output data. For example, when you send text to an AI, the words and punctuation are broken down into tokens, which the AI then processes.

Our token pricing is designed to be simple and predictable. We roll the input and output token costs into a single average value. This consolidated price also covers additional costs, such as Passio API usage, ensuring a consistent billing platform. This approach allows us to update and improve our AI models without needing to change our pricing structure.

The token cost per API call is set to cover the cost of using the service and is priced to align with industry-standard API billing costs. This ensures that our pricing remains competitive and fair within the market.

If you exceed the monthly token limit included in your plan, our billing portal allows you to configure an automatic refill option. By default, this setting is enabled on your account. When enabled, if you make a request to the API and dont have enough tokens, you will automatically be billed at a rate of $2.50 per million tokens to top up the number of tokens you have set.

For example, if your refill amount is set to 2 million tokens, you will be charged $5.00 and gain 2 million more tokens.

There are no volume discounts on refills, but you can manually purchase additional tokens at any time through the dashboard. If the automatic refill option is turned off and you run out of tokens, the service will stop responding until more tokens are added.

‍

Yes, you can upgrade or downgrade your subscription plan at any time to better suit your usage needs. This flexibility allows you to adjust your plan based on changes in your application's user base or feature set.

The REST API is essential for the mobile SDK to gather all necessary food information and is required for the SDK to function. The API offers a wide range of capabilities, including:
‍

• Text-search foods: Easily find foods by searching text.
• Full food details and nutrition: Access comprehensive information about foods and their nutritional content.
• Meal plans: Retrieve and manage meal plans.
• Bar code scan: Scan barcodes to get food information.
• Computer vision routes for food recognition: Use image recognition to identify foods.
• AI assistant tools: Utilize features such as visually similar food suggestions based on user profiles and AI nutritionist conversations.

This comprehensive access ensures that you can fully leverage the capabilities of our food database and Nutrition Advisor in your applications.

‍

Volume discounts are applied based on the number of tokens you use. The rate per million tokens decreases as your usage increases, as follows:
‍

• 1-20M tokens: $2.50 per million tokens
• 20-100M tokens: $2.45 per million tokens
• 100-500M tokens: $2.40 per million tokens
• 500M+ tokens: $2.35 per million tokens

These discounts can significantly reduce your monthly cost if you have high token usage. Volume discounts are not applied on refills.

‍

Our billing portal allows you to configure an automatic refill option, should you exceed the monthly token limit included in your plan. By default, this setting is enabled on your account. When enabled, if you make a request to the API and dont have enough tokens, you will automatically be billed at a rate of $2.50 per million tokens to top up the number of tokens you have set.

For example, if your refill amount is set to 2 million tokens, you will be charged $5.00 and gain 2 million more tokens.

There are no volume discounts on refills, but you can manually purchase additional tokens at any time through the dashboard. If the automatic refill option is turned off and you run out of tokens, the service will stop responding until more tokens are added.

‍

The Mobile app SDK quick-start app is a demo app provided by Passio to give you an example of how you can integrate our modules into your mobile applications quickly and efficiently. It includes sample UI layouts and engineering views so you can peek under the hood at how everything works.

Currently, we do not offer free trials. Upon purchasing, you will be billed immediately for the token credits. This pricing model ensures we can cover the necessary costs required to serve your needs and maintain the high-quality service you expect.

On-device recognition is faster and processes photos on the device itself, meaning the photos are not sent to the cloud. It costs fewer tokens, but only allows you to log a single ingredient at a time, and requires multiple uses to log a meal.

‍

Cloud-side recognition, which uses the Nutrition AI Advisor API, has more tokens per call, but will recognize everything in the image and return all foods, requiring only one main scan to log the meal. It is more powerful but has some latency due to the images being processed in the cloud.

Yes, you can integrate both on-device and cloud-side recognition in your application to leverage the advantages of each method. On-device recognition can be used for quick, simple identification, while cloud-side recognition can handle more complex analysis when needed.

There are no strict limits to the number of users you can have. However, your token usage will increase with the number of users, so you should select a subscription plan that accommodates your user base and expected usage.

Passio Hub takes data security seriously. Data processed via the cloud-side recognition is handled with strict security measures to ensure privacy and protection. On-device recognition keeps the data local to the device, providing an additional layer of security by not transmitting photos to the cloud.

Our standard setup does not store data that can identify the user in any way. Consumers of our SDK might store that data, but Passio does not. The SDK might cache the license key for faster authentication, everything else is just used for the nutritional data. The only component that might be affected is the personalization API we are adding in our 3rd generation SDK that would allow customers to upload end-user profile data (e.g., weight, height, diet, etc.) to personalize responses from the Nutrition AI Advisor.

‍

We do not store this data, but the data is transitioned to our LLM that could be a third party LLM such as GPT. As a technology provider we rely on and require our customers to ensure no personal data is transferred to Passio.