Every Puppy. Every SKU.
The complete One Puppy line — every configuration we offer for reservation, each with a unique SKU, full tech specs, pricing, and the pre-installed 🤫 One Agent. Reserve any for $0.69, fully credited and refundable anytime.
Across the full line
Apple · NVIDIA · Google · Samsung · Microsoft
Shipping configured in 2026
Credited & refundable
Apple·4 SKUs
Unified-memory silicon where the agent runs on the same chip as your life. Day-0 configured, agent live.
One Puppy Max
iPhone 17 Pro Max
The phone becomes your agent. Pocket-sized, always with you.
- SoC
- Apple A19 Pro · 3nm-class
- CPU
- 6-core (2 performance + 4 efficiency)
- GPU
- 6-core Apple GPU, hardware ray tracing
- Neural Engine
- 16-core (TOPS not disclosed by Apple)
- Unified memory
- 12 GB LPDDR5X
- Storage
- up to 2 TB NVMe
- Local models
- 3B–8B int4 (Llama 3.2, Phi-3.5-class)
- Throughput
- ~20–40 tok/s @ 3–8B int4 (target)
One Puppy Max Pro
16″ MacBook Pro · M5 Max
128GB-class on-device AI. The personal supercomputer you own.
- SoC
- Apple M5 Max · 3nm (3rd-gen)
- CPU
- 18-core (6 super + 12 performance)
- GPU
- up to 40-core Apple GPU
- Neural Engine
- 16-core (TOPS not disclosed by Apple)
- Unified memory
- up to 128 GB
- Memory bandwidth
- up to 614 GB/s
- Local models
- up to 70B int4 · 32B at higher precision
- Throughput
- ~10–20 tok/s @ 70B int4 (target)
One Puppy Practice
1× Max Pro + 2× Max · 3 seats
A whole RIA practice, configured, with onsite onboarding.
- Bundle
- 1× MacBook Pro M5 Max + 2× iPhone 17 Pro Max
- Aggregate memory
- 128 GB + 2× 12 GB unified
- Seats
- 3 fully-configured agents
- Onboarding
- Onsite setup + practice playbooks
- Local models
- 70B-class on the Pro, 3–8B on each phone
- Best for
- Wealth practices, small teams
One Puppy · your own device
Your iPhone or Mac
Software only, self-serve, cancel anytime. The easiest way in.
- Runs on
- Apple Silicon Mac (M1+) or iPhone 15 Pro+
- Min memory
- 8 GB unified (16 GB+ recommended)
- Local models
- Scales to your device's unified memory
- Delivery
- Software only — bring your hardware
- Commitment
- Monthly, cancel anytime
NVIDIA·2 SKUs
Grace-Blackwell deskside supercomputing. PetaFLOP-class local inference and fine-tuning. Agent coming to these platforms; reserve your place.
One Puppy Ultra S
DGX Spark · GB10
A petaFLOP on your desk. Personal AI workstation.
- Superchip
- GB10 Grace Blackwell
- CPU
- 20-core Arm (10× Cortex-X925 + 10× A725)
- GPU
- Blackwell · 5th-gen Tensor Cores · FP4
- AI performance
- 1 PFLOP (1000 TOPS) FP4 sparse
- Unified memory
- 128 GB LPDDR5x coherent
- Memory bandwidth
- 273 GB/s
- Networking
- ConnectX-7 · 200 GbE (cluster 2 units)
- Storage
- up to 4 TB NVMe
- Local models
- up to ~200B single · 405B across two units
One Puppy Ultra Max
DGX Station · GB300
Trillion-parameter local models. Datacenter on a desk.
- Superchip
- GB300 Grace Blackwell Ultra
- CPU
- 72-core NVIDIA Grace (Arm Neoverse)
- GPU
- Blackwell Ultra · FP4
- AI performance
- up to 20 PFLOPS FP4
- Coherent memory
- 748 GB (252 GB HBM3e + 496 GB LPDDR5X)
- GPU bandwidth
- HBM3e @ 7.1 TB/s
- Networking
- ConnectX-8 SuperNIC · up to 800 GbE
- Power
- Deskside · standard wall power
- Local models
- Trillion-parameter class · fine-tune 70B locally
Google·1 SKU
Tensor silicon with on-device Gemini Nano. Agent coming to Android; managed infrastructure available first.
One Puppy · Pixel
Pixel 10 Pro XL
Gemini-class on-device AI on Google's own silicon.
- SoC
- Google Tensor G5 · TSMC 3nm
- AI engine
- Next-gen Tensor TPU for on-device Gemini
- Memory
- 16 GB LPDDR5X
- Storage
- up to 1 TB
- Local model
- Gemini Nano (multimodal), 3–4B-class
- Status
- Agent roadmap · managed infra first
Samsung·1 SKU
Flagship Galaxy silicon with a high-TOPS NPU. Agent coming to Android; managed infrastructure available first.
One Puppy · Galaxy
Galaxy S26 Ultra
Galaxy AI hardware with a big on-device NPU.
- SoC
- Snapdragon 8 Elite Gen 5 for Galaxy
- NPU
- Hexagon V81 (TOPS not disclosed)
- Memory
- 12–16 GB (16 GB on 1 TB trim)
- Storage
- up to 1 TB
- Local model
- Personal 3–8B int4 on-device
- Status
- Agent roadmap · managed infra first
Microsoft / Windows·1 SKU
Copilot+ workstations and GB300 desksides for Windows shops. Agent coming to Windows; managed infrastructure available first.
One Puppy · Windows
GB300 deskside / Pro Max towers
Pick a Copilot+ tower or a Grace-Blackwell deskside.
- Deskside option
- NVIDIA GB300 (see Ultra Max specs)
- Workstation NPU
- Copilot+ class · 40–50+ TOPS
- Memory
- up to 128 GB (Copilot+ tower) · up to 784 GB (OEM GB300 workstation)
- GPU
- Discrete RTX / Blackwell options
- Local models
- 8–70B on towers · trillion-param on GB300
- Status
- Agent roadmap · managed infra first
A clean code for every configuration.
Each SKU follows OP-⟨partner⟩-⟨model⟩ — OP for the One Puppy line, a three-letter partner code (APL · NVD · GGL · SAM · MSF), and a short model code. Stable and public, so you, your advisor, or your purchasing agent can reference an exact configuration.
Find your Puppy. Reserve it for $0.69.
Fully credited toward purchase, refundable anytime. We pre-install and configure the 🤫 One Agent before it ships.
One is a product of Hushh Technologies Corporation (brand: 🤫 “hussh”), an independent company. One runs on third-party silicon, systems, and cloud; all company names are used solely to describe the platforms on which One software runs. Hushh Technologies is not affiliated with, endorsed by, sponsored by, or partnered with any company named.