Show HN: New CLI tool to count tokens locally (github.com)

A new command-line interface (CLI) tool has been introduced that enables users to count tokens locally for various Large Language Model (LLM) providers, including OpenAI and Google Gemini. This tool supports a range of models, such as OpenAI's GPT-4.1 and Google's Gemini 1.5, offering the ability to analyze token counts directly from text or files without needing to make API calls—a significant advantage for developers seeking efficiency and resource-saving in their workflows. The tool allows for flexible usage with commands to input text directly or specify files while also enabling users to select different models for tokenization. With features that include fast local execution, exact counting of tokens for supported LLMs, and easy installation options via Homebrew or GitHub, this CLI tool stands to enhance productivity for AI and machine learning practitioners. Its open-source MIT license further encourages community engagement and potential enhancements, making it a valuable asset for those working with token-based models in various applications.