In stock│Windows 10/11 & macOS

Offline AIon a USB.

Name: PortableMind USB — Offline AI for Windows & macOS
Brand: PortableMind
Price: 49.00 USD
Availability: InStock
Rating: 5 (40 reviews)

Plug it into any Windows or Mac laptop. Voice, vision, and chat run locally on your hardware, no cloud, no login, no subscription.

[ ORDER FROM $49 ]How it works →

Three tiers from $49, most pick v1.7.1 ($79). Want offline GPS too? The PRO Navigator adds it for $199.

5.0 · 40 reviews

✓ One-time purchase✓ 30-day returns✓ Ships next business day✓ Secure Stripe checkout✓ 4,000+ units shipped

Now shipping Firmware 1.7.1, Model Manager, Qwen 3.5, multi-modal chat, GPU voice, faster first token. What's new →

US Live Power Gridlive

reading 13 EIA regions…Open full map →

Live US grid pressure from public EIA data. When the grid strains, an offline AI on a USB keeps working. This is the world PortableMind is built for.

Proof

Real people running it

Unscripted demos and reviews from TikTok, not us, them.

▶

Customer demo, @graveyardgoonz

▶

PortableMind clip, @clipstitcher.io

▶

Review, @brandon_veza

▶

Hands-on, @theresalynnreid

Showing 4 of 16 · Real users, real demos

See it running

Real frames from the PortableMind desktop app

Not mockups, not renders, the actual desktop app you get when you plug the USB in.

PORTABLEMIND, DESKTOP APP · OFFLINE

online (local)

Chat

Ask anything, fully offline

›Multiple model sizes, from fast-and-light to high-power
›Saved threads, or an Ephemeral mode that keeps nothing
›One-tap Simplify, Checklist, and Summarize

Real frames from the PortableMind desktop app (footage from 1.5, surfaces unchanged in 1.7.1). Click the tabs to look around.

Open the full demo, tour + grid-down answers →

New · Flagship

Meet the PRO Navigator

Everything PortableMind does, chat, voice, vision, now with a full offline US GPS: turn-by-turn routing and ~258M addresses, working with zero signal. Ships with a USB GPS receiver in the box.

Explore the Navigator →Order $199

● New flagship✓ Windows 10/11✓ 100% offline

PortableMind PRO Navigator kit, the Navigator USB in a slide-tin case, a BU-353N5 USB GPS receiver, and a 100W USB-C adapter laid out on a field desk beside a compass and topo map

Compatibility check

Will it run on your machine?

Two questions, a straight answer. We'll tell you if it runs and which tier fits, honestly, even if the answer is “check with us first.”

Your computer

Memory (RAM)

Pick your OS and RAM above.

Firmware 1.7.1 (current build)│In stock · ships next business day

Three tiers. One offline AI.

Pick the tier that matches your needs. All include Firmware 1.7.1 with voice, multi-modal chat, SCOUT Vision, phone access, and the in-app Model Manager. Windows and macOS support available on v1.7.1 and MAX-SPEED.

4,000+units deployed

CORE

$49one-time

Windows

Windows 10/11
Preloaded model roster
Chat mode
AES-256 encryption

[ ORDER CORE ]

v1.7.1

$79one-time

WindowsmacOS

Windows + macOS
Voice mode
Multi-modal chat + SCOUT
Model Manager

[ ORDER v1.7.1 ]

MAX-SPEED v1.7.1

$129one-time

WindowsmacOS

Everything in v1.7.1
400 MB/s drive
Faster boot + switching
Windows + macOS

[ ORDER MAX-SPEED ]

// Secure checkout via Stripe. Taxes and shipping calculated at checkout.

New · Flagship

PRO Navigator, offline AI + GPS in one kit

Turn-by-turn routing with zero signal · $199

// Operator feedback

5.0 · 40 reviews

Real feedback from operators running PortableMind offline in the field, at home, and in the library.

See all reviews →

“this usb has exceeded all my expectations!”

Avery H.

“this thing is legit!”

Riley B.

Why offline

AI that doesn't depend on a connection

No cloud, no outage

Cloud AI dies the moment the signal does. PortableMind runs on the machine in front of you, on a plane, off-grid, or in the middle of an outage.

Private by default

Your conversations, documents, and voice never leave your hardware. No telemetry, no uploads, no account to create.

Pay once, own it

One USB, one payment. No subscription to cancel, no model quietly nerfed. It keeps working whether or not you ever update again.

Built forTravelersPrivacy-focused professionalsOutage & emergency prepField & remote workWriters & researchers

On every drive

Voice, vision, and a real desktop app

Firmware 1.7.1 ships with the in-app Model Manager, the full Qwen 3.5 model roster, multi-modal chat (images in any turn, not just SCOUT), GPU-accelerated voice on Windows, and the fine-tuning pass that made first-token latency 10 to 30x faster on tight VRAM. All local, all private.

Full firmware tour →

Your USB, your rules

Model Manager

Download any Ollama-registry model onto the drive from inside the app. Delete models with one click. See disk usage. Get VRAM-aware recommendations for what will actually run on your hardware.

In-app overlay lists every installed model, size, modified date, and which is active.

Streaming progress bar for downloads, one-click delete with confirm.

Curated catalog with a new Light 1.5B preset for very old or very cheap machines.

A whole new model family

Qwen 3.5 Roster

Every preset moved to Qwen 3.5. Better reasoning, tighter answers, faster first-token latency across Fast 2B, Smart Default 4B, Smart Premium 9B, and Scout 4B vision.

Fast, Smart Default, Smart Premium, and Scout all on Qwen 3.5.

Experimental alternative Scout on Gemma3:4B for A/B vision comparisons.

Auto-installed on first launch; existing chat history stays intact.

Images in any turn

Multi-modal Chat

Drop an image into any chat turn and the model reads it. Vision is no longer locked to the SCOUT tab; the whole chat surface is image-aware if the routed model can see.

Drag/drop or paste an image directly into the chat window.

SCOUT tab still exists as a dedicated image workspace.

Router transparently sends images to the vision-capable model.

3 to 6x faster on Windows

GPU-Accelerated Voice

Kokoro TTS moved to DirectML. First launch auto-converts the ONNX opset. Voice replies drop from 8 to 15 seconds down to 2 to 4 seconds on compatible Windows GPUs.

DirectML-accelerated on Windows with a compatible GPU.

Auto-fallback ladder: q8+DML then fp16+DML then q8+CPU.

Piper remains the voice stack on macOS.

No more VRAM swap penalty

Unified Voice + Chat Residency

On cards below 10 GB VRAM, voice reuses whatever model chat is running. Zero swap penalty when you switch between them. On larger GPUs, voice keeps its own dedicated model for parallel throughput.

Automatic based on your real card, no config needed.

Removes the 6 second+ VRAM swap that used to hit voice on tight hardware.

Live re-detection if you plug in an eGPU or update drivers.

Faster, quieter, harder

Fine-Tuning Pass

First-token latency dropped from 60 to 90 seconds down to 3 to 8 seconds on 6 GB VRAM with the 9B Premium model. Memory decays cleanly. Transient inference errors self-recover.

KV cache sized per model tier, no more spillover thrash.

Taught facts decay weekly if not reinforced, protected fact types decay slower.

2 automatic retries with backoff on transient EOF errors.

Still on the drive

Everything from 1.5

Voice mode, SCOUT Vision, Phone Access, the desktop-app launcher, thread history, AES-hardened import/export, and the offline Field Manual all remain intact. Nothing was removed to make room.

Voice mode, SCOUT Vision, Phone Access, desktop-app launcher.

Thread history, AES-256 import/export, Offline Field Manual.

Cross-platform: Windows 10/11 and macOS (Intel + Apple Silicon).

Built-in field manual

Six categories of grid-down reference ship on every drive, browsable offline with zero AI calls. Tap Ask AI about this when you want the model to walk you through your exact situation.

Water

Fire

First aid

Shelter

Signal

Food

Want a permanent desk setup? The Survival Terminal is the same offline brain, hand-built into a dedicated box. Released in small runs.

View Terminal →

FAQ

Common questions

Does it really work offline?+

Yes. The AI runs locally on your Windows 10/11 or macOS machine, no account, no cloud API key, no connection. Internet is only needed if you choose to download an optional update later.

What do I need to run it?+

A reasonably modern Windows 10/11 PC or Mac with at least 8 GB of RAM and a USB port. USB-C works fine with an adapter.

Is there a subscription or API key?+

No. You pay once for the PortableMind USB. The models run on your hardware with no recurring fees and no cloud keys.

What can it actually do?+

Chat, voice conversations, image and document recognition (SCOUT Vision), and phone access, all running locally on your machine with no internet required.

What's the difference between the tiers?+

CORE is the $49 Windows entry point: curated Qwen 3.5 roster, chat, AES-256. v1.7.1 ($79) adds voice, multi-modal chat, SCOUT Vision, in-app Model Manager, phone access, and macOS. MAX-SPEED v1.7.1 ($129) is v1.7.1 on a 400 MB/s drive for faster boots and model switching.

Free field guide│Print-ready PDF

The Offline Readiness Field Guide.

Six fact-checked sections covering power, water, comms, offline AI, navigation, and first aid, with the exact numbers most people get wrong and the survival myths that get people hurt. Sourced to CDC, EPA, FEMA & NIST, and yours as a polished, print-ready PDF.

Free PDF, plus firmware drops and the occasional subscriber-only discount. No spam, unsubscribe anytime.

Offline AI. Ready the moment you plug in.

One USB, one payment. Voice, vision, chat, and phone access, running locally on your hardware, with no cloud to depend on.

[ ORDER FROM $49 ]

✓ From $49✓ Ships next business day✓ 30-day returns✓ Secure Stripe checkout✓ 4,000+ units shipped