Username Studio addresses the difficulty of finding available Telegram usernames by providing a local tool. The tool leverages a local large language model (LLM) to generate, score, and check short usernames on your computer. It offers three username generation styles: brandable, Russian transliteration, and multilingual. Usernames are scored based on readability, brandability, meaning, and rarity before being filtered. The tool then optionally checks availability on Telegram using the Telethon library. All generated usernames and their statuses are saved locally using SQLite. A local web dashboard, accessible via a web browser, provides an interface to browse, score, and check usernames. A CLI mode is available for users who prefer working in the terminal. The tool prioritizes safety with options to disable Telegram connections and preview actions. A Windows-specific quick start guide is provided for easy setup and usage. Even without LM Studio, the tool provides fallback generation and scoring capabilities. Future developments include Linux/macOS support more generation styles, and configurable username length filters. The project's GitHub repository is provided to encourage feedback and collaboration. The project creator invites feedback from users, particularly regarding improvements to the scoring process.