tokenwise

Estimate prompt token usage, compare model costs, and optimize prompts directly in VS Code with live token counting and instant analysis.