Tracking calories is one of the biggest problems in personal health. This is because most people give up on manually logging their food due to the inconvenience of the process. In order to address this issue, the paper presented here is about a web-based calorie calculator that implements Google’s Gemini 2.0 Flash multimodal AI for an automatic food identification and nutritional value estimation from images. The system employing user-centered design and deep learning to be quick and accurate in the tracking process has been developed with a React frontend and Next.js/Edge Runtime backend. It can recognize food with more than 85% accuracy and has a response time of 2–5 s. The article details the system architecture, the processing flow, the error handling, and the use of the system in wellness programs, personal diets, and clinical nutrition.