AI Spend refers to the budget allocated for your bot's usage of Large Language Models (LLMs), which are used in tasks like generating responses and processing natural language inputs. Each action your bot performs that involves AI consumes tokens, and these tokens have associated costs set by the model providers.

Botpress passes these costs directly to you without any markup, meaning you’re getting the exact same price for the same number of tokens if you were to work directly with a model provider.

To make sure you can get started building and prorotyping properly, we provide a free monthly credit of $5 towards AI Spend. If your bot exceeds this credit, you’ll need to increase your AI Spend limit to maintain uninterrupted service. You can adjust this limit by increasing your AI Spend limit in your workspace dashboard. AI Spend on Pay-as-you-go plans cannot currently exceed $100—if you need a higher amount than this, feel free to get in touch with us using our website.

Monitoring your AI Spend is important for budgeting efficiently and ensuring your bot operates within its allocated resources. When building your bot, the event debugger has detailed information about token consumption broken down by action. Once your bot is live, you can check your production logs to better understand token consumption. The Analytics tab for your bot also has detailed information about LLM performance, cost, and token consumption.