Local GGUF AI Copilot & Live Omnibox Calculator

Run offline llama.cpp GGUF models on Apple Silicon Metal or evaluate math, currencies, and CSS units in search.

Local GGUF AI Copilot & Live Omnibox Engine

Sulguni Cortex combines an offline-capable AI Copilot with an instant Omnibox command engine.

Local llama.cpp GGUF vs Cloud BYOK

In Preferences → AI Copilot, choose how Cortex processes AI transformations:

  • 100% On-Device Local GGUF (llama.cpp Metal): Download quantized GGUF models (Qwen-2.5-3B-Instruct, Llama-3.2-3B, Mistral-7B) and run inference directly on Apple Silicon GPU with zero internet traffic.
  • Cloud BYOK: Connect your own API key for OpenAI, Anthropic Claude, Google Gemini, or a local Ollama server.

Built-In 1-Click AI Actions

  • Summarize Key Points & TL;DR
  • Extract Action Checklist
  • Refactor & Explain Code
  • Rewrite Executive Reply
  • Strip Formatting & Redact PII

Live Omnibox Calculator, Currency & Unit Converter

Type any expression directly into the Cortex search bar for sub-millisecond evaluation:

  • Math: (1490 * 12) * 0.82 → 14,661.6
  • Currency & Crypto: $240 in EUR, 0.25 BTC to USD
  • CSS & Data Units: 48px to rem, 512 MB to GB, 72 F to C
  • Hex / Binary: 0xFF5A00 → 16734720 (DEC)