{"product_id":"haven","title":"Haven","description":"\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003e\u003cstrong\u003eHaven\u003c\/strong\u003e is a private AI assistant that lives on your desktop — not in someone else's data center.\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003eEvery word you say, every email it reads, every fact it remembers stays on your machine. No cloud. No API keys. No telemetry. Ever.\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003eHaven runs its own language models locally via GGUF inference, automatically selecting the best model for your hardware — from a 1B parameter model on 8GB RAM machines all the way to 14B on high-end GPUs. It speaks, listens, and thinks without sending a single byte off your PC.\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003e\u003cstrong\u003eWhat Haven does:\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003eTalk to it by voice or text. Haven manages your email, calendar, contacts, files, and desktop — anticipating what you need before you ask. It monitors your schedule, watches for urgent messages, and delivers morning briefings and end-of-day summaries on its own. It controls your smart home, launches your apps, searches the web privately through DuckDuckGo, and learns your patterns over time.\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003eUnder the surface, Haven runs a 7-stage AI command pipeline, a knowledge graph that builds a model of your world, hybrid semantic search across your documents, and 14 proactive triggers that surface the right information at the right moment.\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003e\u003cstrong\u003eWhat you see:\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003eA living sphere rendered in real-time OpenGL — a breathing, color-shifting neural constellation that reflects Haven's state. No chrome. No clutter. Just a presence on your desktop that's always ready.\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003e\u003cstrong\u003eBuilt for your hardware:\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003eHaven's 6-tier system auto-detects your GPU and RAM and configures everything — LLM model, voice engine, context window — to run at peak performance on your specific machine.\u003c\/p\u003e\n\u003cul class=\"[li_\u0026amp;]:mb-0 [li_\u0026amp;]:mt-1 [li_\u0026amp;]:gap-1 [\u0026amp;:not(:last-child)_ul]:pb-1 [\u0026amp;:not(:last-child)_ol]:pb-1 list-disc flex flex-col gap-1 pl-8 mb-3\"\u003e\n\u003cli class=\"whitespace-normal break-words pl-2\"\u003e\n\u003cstrong\u003eGPU Elite \/ High\u003c\/strong\u003e (12–16 GB+ VRAM) — Qwen 2.5 14B, Kokoro TTS, faster-whisper STT\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words pl-2\"\u003e\n\u003cstrong\u003eGPU Mid\u003c\/strong\u003e (8 GB VRAM) — Llama 3.1 8B, full voice pipeline\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words pl-2\"\u003e\n\u003cstrong\u003eGPU Entry\u003c\/strong\u003e (4–6 GB VRAM) — Gemma 3 4B, GPU-accelerated\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words pl-2\"\u003e\n\u003cstrong\u003eCPU Strong\u003c\/strong\u003e (16 GB+ RAM) — Gemma 3 4B, CPU inference\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words pl-2\"\u003e\n\u003cstrong\u003eCPU Basic\u003c\/strong\u003e (8 GB RAM) — Gemma 3 1B, lightweight voice via Piper + Vosk\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003e\u003cstrong\u003eSystem requirements:\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003eWindows 10 or 11. Python 3.12. 8 GB RAM minimum. GPU optional but recommended. No internet connection required after initial setup.\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003e\u003cstrong\u003eWhat's included:\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp class=\"font-claude-response-body break-words whitespace-normal leading-[1.7]\"\u003eOne-time purchase. Full Haven 1.0 licence. All future 1.x updates included.\u003c\/p\u003e","brand":"Emphos Group","offers":[{"title":"Default Title","offer_id":50662768017655,"sku":null,"price":79.99,"currency_code":"USD","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0931\/8409\/8551\/files\/Haven_AIforthefutureofwork.png?v=1775530632","url":"https:\/\/emphosgroup.com\/products\/haven","provider":"Emphos Group","version":"1.0","type":"link"}