Sal. A private partner for your Mac.

Sal is not a cloud AI running on your laptop. It is designed to run entirely on your Mac — embracing what that makes possible. Because Sal lives locally, she can read your Calendar, search your files by meaning, glance at your screen when you invite her, and remember only what you pin. She is tightly integrated with your work environment while keeping everything private and secure. Cloud assistants cannot do this without transmitting your data to remote servers.

Nearly every AI assistant built for Mac users today is designed for enthusiasts — people comfortable choosing model quantizations, configuring runtimes, and debugging GPU memory errors. Sal was engineered for the other 99%. By embracing tight macOS integration instead of pretending to be a cloud service running elsewhere, HelloPartner.ai made privacy and depth non-negotiable from day one.

• 01

  Completely Private, by Architecture

  Sal does not "phone home." She does not collect telemetry, analyze usage, or send any data to HelloPartner.ai or any third party. She is mechanically incapable of it — there is no cloud endpoint to send data to. Your conversations, files, and documents stay on your Mac. Full stop.

• 02

  Accessible Monthly Pricing. No Token Taxes.

  Sal costs just $4.99/month for Mac (with a 14-day free trial included) and $2.99/month for iPhone (currently free during the TestFlight public beta). You own your AI experience — no usage tiers, no API token taxes, and no compute metering. Because Sal runs on hardware you already own — Apple Silicon — the compute cost per query is effectively zero. HelloPartner.ai passes that savings directly to you.

• 03

  Mac-Native Intelligence — Not a Chat Window in the Cloud

  Sal is a real macOS application. She reads the macOS Calendar and Reminders databases directly. She indexes your local file system at the semantic level — understanding what your documents mean, not just what they're named. And when you want her to see your screen, you invite her explicitly — with a keyboard shortcut or a force-press of the PTT button — and she glances, responds, and immediately forgets. Your desktop is never monitored passively. No other AI assistant integrates with the Mac at this depth while respecting this level of privacy.

• 04

  A Custom Inference Engine Built Directly on Apple MLX

  To our knowledge, Sal is the only consumer AI product that has built its own inference engine directly on top of Apple's MLX framework — the same low-level GPU/Neural Engine libraries Apple engineers use internally. This was not a trivial undertaking. It required months of model evaluation, parameter tuning, memory profiling, and response-quality benchmarking across Apple Silicon variants.

  The payoff is significant: all of Sal's capabilities — language, vision, voice, image generation — run through a single centralized engine. There is no mix of competing runtimes to install, no separate app for each task, and no memory allocation conflicts across processes. It is the architectural foundation that makes a truly zero-setup consumer experience possible. The specific models Sal uses are available to verified press upon request.

• 05

  Works Without Wi-Fi

  Because Sal is entirely on-device, she works anywhere. On planes. In remote cabins. In buildings with no signal. For a knowledge worker, the ability to think alongside an AI collaborator in any context — not just when a server can be reached — changes how work gets done.

• 06

  Zero Setup. Every Decision Already Made For You.

  Running a local AI assistant without Sal means choosing a model from a catalog of hundreds — each with a name that reads like a firmware revision — then installing separate runtime software for chat, vision, image generation, and voice, then configuring memory limits manually, and hoping the combination doesn't crash. Most people give up before the first response.

  HelloPartner.ai spent months making every one of those decisions — model selection, quantization depth, memory scheduling, response quality thresholds — and then built a unified engine so those decisions never surface to the user. Download the app, enter a license key, and Sal is ready to work. That is the entire setup process.

• 07

  The Pinning System — Transparent Memory You Can See and Control

  Local hardware imposes real constraints. Sal cannot accumulate an unlimited number of documents the way a cloud AI can — and HelloPartner.ai's response was not to apologize for this, but to ask a harder question: do users actually need every document in context at all times? After months of research and iteration, the answer was no. Most users need the documents that are in focus right now.

  From this insight came the Pinning System. Users drag the documents they're actively working with into Sal's Pinned panel. Sal treats them as the center of attention. User and Sal are looking at exactly the same things, at exactly the same time. When a document is no longer needed, unpinning it removes it from Sal's working memory — freeing space and sharpening focus. There is no guesswork. The user opens the message history and what they see is precisely what Sal sees. What you see is what she knows.

  This level of transparency is structurally impossible for a cloud AI. A remote server accumulating your files cannot offer the user genuine visibility into what the model is holding, or genuine control over when it forgets. Sal can — because she lives entirely on your Mac, and her memory is entirely yours to see.

Sal's primary interface — a focused dark-mode chat window that lives in your menu bar and appears exactly where you need it.
↓ Download full-resolution image

Semantic file search: Sal finds documents by what they mean, not what they're named.
↓ Download

Calendar integration: Sal reads and writes to the macOS Calendar app directly, with no cloud relay.
↓ Download

Reminders integration: Sal writes tasks directly to macOS Reminders — no third-party sync, no cloud. Already on your iPhone.
↓ Download

Ephemeral Sight: Press Shift+Cmd+Return (or force-press PTT while dictating) to invite Sal to glance at your screen. She sees, responds, and forgets. Your desktop is never monitored passively.
↓ Download

Document awareness: Sal reads a pinned PDF and answers questions about it — while you keep working in Finder. Everything stays on your Mac.
↓ Download

Personalities: Sal ships with multiple built-in character profiles, each fully rebrandable.
↓ Download

All screenshots are approved for editorial use with attribution to HelloPartner.ai. High-resolution originals are available upon request. Do not crop, composite, or alter screenshots beyond standard color correction.

Download hi-res headshot  ↓

Neil Balthaser is a 20-year veteran of enterprise software who loves creating innovative products that change the world. He founded Intellogo (acquired in 2018), where he pioneered the “Conversation as a Platform” model – connecting developers to end users through artificial intelligence and machine learning. He also held senior level executive positions at Adobe and Barnes & Noble. A builder by nature, Neil has seen every wave of this industry: chatbots, voice assistants, cloud-based AI services. A solo tech founder and strategist, Neil built Sal because he needed a local AI assistant with her capabilities badly enough to code it. As such, he’s fully committed to building a local AI assistant for knowledge workers built with compute sovereignty and professional utility in mind.