Standalone client-side calculators and configuration builders for AI engineering workflows. Each tool runs on its own dedicated page so you can bookmark, share, or search specific calculators directly.
Reference: Anthropic · Google DocsModel monthly LLM API token bills across Claude Fable 5.1, Opus 5.5, Sonnet 5.5, Haiku 5.5, and Gemini 4 Argon with 5-minute/1-hour prompt caching and 50% async batch discounts.
Reference: Hugging Face · Ollama DocsCalculate static GGUF weight memory (Q4_K_M, Q5_K_M, Q8_0, FP16), architecture-based FP16 KV cache across 4K–128K context windows, and GPU capacity targets for 8B–70B models.
Reference: MCP · Node.js DocsGenerate editable claude_desktop_config.json templates with Windows cmd.exe /c npx wrappers (fixing spawn npx ENOENT), macOS Homebrew/nvm absolute paths, and Python uvx launch settings.