Fiorino AI is an open-source platform that addresses a growing challenge for SaaS businesses: understanding and controlling the costs of integrating large language models (LLMs) into their products. As more companies build AI-powered features using providers like OpenAI, Anthropic, and others, tracking exactly how much each user, feature, or subscription tier spends becomes increasingly complex. Fiorino AI tackles this by providing granular visibility into AI spending at the per-user level, complete with configurable spending limits and automated billing workflows.
Core Features
At its foundation, Fiorino AI acts as a cost-tracking middleware between your application and the LLM providers you use. It processes every message your users send and every response the AI generates, logging token usage and calculating costs based on a configurable database of model pricing. You can set custom overhead margins to account for infrastructure and operational expenses, giving you a true picture of your AI costs rather than just raw provider charges.
One of the platform's standout capabilities is per-user usage monitoring. For SaaS businesses with subscription plans, this means you can track AI consumption on a per-subscriber basis, receive alerts when users approach their plan limits, and optimize your pricing tiers based on real usage data. For pay-as-you-go services, Fiorino AI integrates with Stripe to bill users automatically based on their exact AI usage, with the ability to set custom margins and manage overhead costs efficiently.
The platform also offers an AI Cost Analyst feature, which lets you chat with your usage data in natural language. Instead of combing through dashboards, you can ask questions about cost patterns, receive optimization suggestions, and identify usage trends through a conversational interface. This transforms raw metrics into actionable business intelligence without requiring deep data analysis skills.
Integration and Deployment
Fiorino AI is designed for flexible deployment. You can run it using Docker with a single command, connecting it to your PostgreSQL database. The platform provides both a web dashboard for visual monitoring and a REST API for programmatic access. Integration with your application happens through its SDK or API, requiring just a few lines of code to start tracking all AI interactions. The platform also supports Stripe for direct user billing, creating a complete loop from usage tracking to revenue collection.
The open-source nature of Fiorino AI means you retain full control over your data and deployment. The project is licensed under Apache 2.0 and welcomes community contributions. Its scalable architecture uses anonymous user IDs to protect privacy while maintaining accurate tracking, and it provides comprehensive audit trails for compliance purposes.
Use Cases
Fiorino AI serves several distinct audiences. SaaS businesses with subscription-based pricing can monitor AI usage per subscriber to ensure they stay within plan limits, receiving proactive alerts when users approach their quota. Pay-as-you-go services can track exact costs per user and automate billing through Stripe, managing overhead costs with precision. AI-powered applications, including chatbots, content generators, and any product using LLMs, can gain a clear understanding of their cost structure and optimize their pricing strategy accordingly.
By providing this level of visibility and control, Fiorino AI helps SaaS businesses move from guessing at their AI costs to managing them with confidence, ultimately protecting profit margins as AI usage scales.
