Versions:
AI Consumption Tracker 2.3.3, published by Alexander Brandt, is a lightweight system utility designed to log and visualize token expenditure across local and cloud-based large-language-model sessions. Positioned in the Diagnostics & Benchmarking subcategory, the application parses JSON or CSV exports from OpenAI, Anthropic, Azure, and self-hosted endpoints, then renders daily, weekly, and project-level breakdowns of prompt, completion, and cached-token costs. Network administrators use it to enforce departmental budgets, while freelance prompt engineers rely on its per-client ledger to generate precise invoices; enterprise architects employ the comparative dashboard when stress-testing competing models before wider rollout. Since its initial release four versions ago, the program has evolved from a command-line script into a portable Windows executable that requires no elevated rights, stores its SQLite database in the user profile, and can auto-purge records older than a configurable retention period. Version 2.3.3 introduces support for multi-currency billing zones, real-time tray alerts when an API key approaches a defined threshold, and a dark-mode report viewer that exports directly to PDF or Excel. Earlier iterations (1.8, 2.0, 2.1, 2.2) remain available for legacy Windows 7 environments, each archived with matching checksums on the project page. The software is available for free on get.nero.com, with downloads provided via trusted Windows package sources (e.g. winget), always delivering the latest version, and supporting batch installation of multiple applications.
Tags: