PHNTM One specifications

A private AI appliance. Every part, listed.

PHNTM One combines a Raspberry Pi 5 with 8 GB RAM, a 10.1-inch touchscreen and local conversation, memory, document search and voice. For people who want a dedicated assistant with Private mode on their own hardware. $799; no required subscription.

Raspberry Pi 5 · 8 GB10.1″ touchDebian 13Gemma 3 4B on-device

Prefer a working example? Inspect the measurements or read the Raspberry Pi build log.

8GBRAM, budgeted at boot for the model; after a long idle it unloads and reloads on your next question
10.1″1920×1200 IPS touchscreen, 5-point touch
256GBstorage for local models, memories and your files
4Bparameter model (Gemma 3), running on the device
27WUSB-C power supply; no battery
0telemetry, accounts or PHNTM servers
What's inside

Five layers. Nothing hidden.

Scroll and it comes apart. Every part is a real, off-the-shelf component you can look up, and the full bill of materials, with prices, is public.

  1. Glass and touch5-point multi-touch over the whole screen. Setup, chat and everything else happen right here.
  2. 10.1″ IPS display1920×1200. The breathing orb, the live status, the conversation.
  3. Raspberry Pi 5 · 8 GBQuad-core Cortex-A76. Runs Debian 13, the model, memory, voice and search.
  4. Active fanGoverned by the Pi's firmware: quiet (often off) at idle, it ramps under load. Low-50s °C sustained, no throttling observed.
  5. Enclosure, speaker and storageA built-in speaker so it can talk back, and an included 256 GB NVMe M.2 SSD.
Compute

Raspberry Pi 5

8 GB RAM · quad-core Cortex-A76

Powerful enough to run a real language model entirely on the device, and small enough to sit on your desk. A boot-time RAM budget keeps the chat model loaded; under pressure it drops to a smaller fallback model and says so.

Display

10.1″ touchscreen

1920 × 1200 IPS · 5-point touch

Set it up and talk to it right on the screen. No computer is needed to get started, ever.

Storage

256 GB, included

NVMe M.2 SSD

Storage for local models, your conversations and documents. Offline reference packs are planned 1.0 additions, not yet release-verified. NVMe storage is included in all newly shipped devices.

Voice

Speaker + push-to-talk mic

tap-to-talk by default · optional local wake detection

A built-in speaker so it can talk back, and a USB mic that's live only while the button blinks red. Listening and speaking both run on the device.

Software

Debian 13, AI built in

local model · zero telemetry

A full Debian system with the assistant built in. SSH is documented and never removed, and the code ships as readable source on the device.

The assistant

Private by default

three modes · you choose

A capable on-device model, free forever. Optional Boosted mode uses Claude on your own Anthropic key, and every answer is labelled with the brain that produced it. Optional Autonomous mode adds actions from your own Telegram, and every email send or calendar change waits for your yes.

Full technical sheet

For the people who read spec sheets.

The same facts as the manual that ships on the device, where every unit also serves its own burn-in certificate at /docs/certificate. What it all adds up to is on the features page.

HARDWARE
compute
Raspberry Pi 5 · 8 GB RAM · quad-core Cortex-A76
display
10.1″ 1920×1200 IPS · 5-point touch (HDMI panel)
audio out
Built-in speaker
microphone
USB mic, included · push-to-talk only
keyboard
Mini wireless keyboard + thumb touchpad, included · TAB = talk
storage
256 GB NVMe M.2 SSD · included
network
Gigabit Ethernet · Wi-Fi 2.4/5 GHz
power
27 W USB-C PD supply, included · no battery, no clock battery
cooling
Active fan, firmware-governed
SOFTWARE & AI
os
Debian 13 (trixie) · stock kernel · systemd
chat model
Gemma 3 4B · Q4_K_M · 8K context
fallback
Llama 3.2 3B · Q4_K_M · used under RAM pressure
embeddings
nomic-embed-text · local memory and document search
speech in
whisper.cpp · on-device · runs per utterance
speech out
Piper TTS · on-device
inference
Ollama / llama.cpp · localhost only · run with --offline
optional
Boosted mode · Claude on your own Anthropic key, off by default · Autonomous mode · the same, plus confirmed email and calendar actions from your Telegram
WHAT IT DOES
voice
tap-to-talk · Notes clips up to 60 s · planned read-aloud
languages
Documented · local Text replies support ten language choices. Speech in all ten languages remains planned for 1.0, not yet release-verified. Do not assume a text-language choice provides the same spoken support.
your day
Documented · local Save one-off or recurring reminders on the device. Today’s accepted fixes preserve pause state, recurring schedules across timezone changes and pending delivery records. Alerts still depend on a verified clock. No RTC coin cell is installed, so an offline cold power loss can leave wall-clock time unknown. Timely offline alerts after that power loss are not established.
read & know
PDF, text, Markdown · photos and scanned PDFs (on-device OCR) · cited answers
offline library
Planned · not release-verified Planned 1.0 capability, not yet release-verified: local Wikipedia and reference packs with source-linked answers. The current local model can answer general questions, but that is separate from a reference library.
smart home
Home Assistant over your LAN · lights, plugs, scenes, climate · confirm for locks
skills
sandboxed · no network · no credentials · owner-approved
self-care
bounded repair paths · planned weekly report
HISTORICAL PROTOTYPE MEASUREMENTS
speed
about two words a second · a short answer usually takes tens of seconds
first answer after idle
slower, once, while the model reloads
power to ready
August prototype: about a minute; current retail boot timing is not established here
temperature
August v0.8/NVMe prototype: low-50s °C under load · no throttling observed in that test
NETWORK & PRIVACY
open ports
22 (SSH), 8843 (app, HTTPS) and 8800 (redirect only), private LAN only · mDNS
model server
bound to the device itself; the firewall drops its outside traffic
telemetry
none · no account · no PHNTM servers
updates
you tap Install · signature and every file checked · self-test gated · automatic rollback
In the box

What you unpack.

  • PHNTM One: Pi 5 inside a 10.1″ touchscreen case
  • USB microphone for push-to-talk
  • Mini wireless keyboard with thumb touchpad
  • 27 W USB-C power supply (US plug)
  • 256 GB NVMe M.2 SSD, everything installed
  • Printed quick-start card and a card with your unit's serial (its burn-in certificate is on the device)
Ready when you are

Own your AI. Keep your words.

Order your PHNTM One. Every unit is built by hand and burned in for at least 24 hours before it ships. $799 once, with no required subscription for local use.

$799 one-time · ships within 7–10 business days after payment · free US shipping · 30-day returns

$799 once · no subscription
Order now