PLANNED
VRAM Calculator
Plan weight memory and runtime overhead.
Read the related guide →INFERENCE LAB / TOOLS
Tools
Small, focused utilities for planning LLM deployments.
The tools below are planned. Until their assumptions are validated, use the linked engineering notes to guide your own measurements.
PLANNED
KV Cache Calculator
Estimate attention state from model geometry.
Read the related guide →PLANNED
Context Length Calculator
Understand context, concurrency and cache capacity.
Read the related guide →PLANNED
vLLM Command Generator
Build version-aware serving configurations.
Read the related guide →PLANNED
SGLang Command Generator
Capture model-specific deployment requirements.
Read the related guide →