Rig

rig.ai/

Rig is a coding assistant designed to run entirely on-device — no cloud API calls, no usage caps, no telemetry — using an 8GB model built specifically for engineering tasks rather than a general-purpose LLM. Its headline reads "On-device AI coding. No cloud. No limits," and it backs the local-first pitch with concrete hardware specs: it requires Apple Silicon M2 or later with 32GB RAM, claims a 300ms first-token latency versus 400-1,000ms for cloud APIs, and states the model dedicates 100% of its parameters to coding versus roughly 15-20% for general models. Notably, the page is upfront about the product's immaturity, stating plainly "Rig's model is still in development so we do not have benchmarks available yet," with a Q3 2026 release target and early-access signup rather than a shipped product.

Category

Developer Tools

Style

DarkColorful

Fonts

ChaletInstrument SansChivo Mono

Copy CSS

Palette

Copy allCopy CSS vars

Meta title

Rig - Local-First AI Coding Assistant

Sections

5

Updated

Sep 22, 2026