Tools

Useful, explainable tools.

Small technical tools with visible assumptions, documented methodology and links back to the concepts behind the calculation.

Local AI · New interactive tool

Local Model Fit Checker

Combine model size, context, GPU VRAM and system RAM into an explainable planning signal for GPU fit, partial offload or RAM-heavy execution.

Check a configuration →
Local AI · Interactive

Local LLM VRAM Calculator

Estimate model weights, architecture-aware KV cache and a configurable runtime reserve—without pretending one number can guarantee GPU fit.

Open calculator →
AI Models · Interactive

Model Size & Memory Explorer

Compare theoretical weight payloads across precision classes and optionally add a conventional KV-cache estimate.

Open explorer →