Message credits system

Our platform operates on a message credits system, where each query posed by a user consumes a certain number of message credits. This system is designed to provide flexibility and efficiency, accommodating the diverse computational needs of different AI models.

Credits Required per Model

Each AI model available on our platform requires a different amount of message credits per query. This structure allows users to select the model that best fits their requirements, balancing between the depth of insight needed and the credits they wish to expend.

Why the Difference in Credit Cost?

Our AI model offerings are designed to balance accuracy, response time, and computational demand, providing flexibility to meet diverse user needs.

GPT 4o-mini is our base model, delivering fast and accurate responses at a cost-effective price. It is ideal for applications where speed is a priority without compromising too much on accuracy.

GPT 4o is our top recommendation for new projects. It offers the highest accuracy, comparable to the previously recommended GPT 4, but operates with greater efficiency, making it twice as fast as GPT 4 Turbo and significantly lower costs. GPT 4o is the preferred choice for users seeking a combination of speed and in-depth, accurate insights.

We no longer recommend GPT 4 and GPT 4 Turbo due to their higher costs and inferior performance metrics compared to GPT 4o. However, these models remain available for users who prefer to maintain the behavior of their existing chatbots.

Additionally, GPT 3.5 Turbo has been deprecated and is no longer available for selection in new projects. However, ongoing projects running on GPT 3.5 Turbo will continue to do so to avoid altering the behavior of active deployments.

By choosing between GPT 4o and GPT 4o-mini, users can effectively tailor their experience on our platform according to their specific requirements, whether they prioritize maximum accuracy with GPT 4o or cost-effectiveness with GPT 4o-mini.

Algorithm cost

In addition to the base model costs, we offer the option to enhance your queries with the fusion algorithm. This add-on helps to refine the responses further but requires additional computational resources.

Please note that the additional credits for the fusion algorithm are added on top of the base credits required for the selected model.