Nano Pro Configure
An engineering team at work in a bright open-plan office, with a Renderboxes Nano Pro workstation on castors beside the desk
A Renderboxes Nano Pro workstation on castors beside an oak desk bank in an open-plan office

Four GPUs.
Under your
desk.

Nano Pro, Maximum power, minimum footprint.

Nano Pro

Compact outside, colossal inside.

Nano Pro packs workstation-class compute into a compact desk-side machine. It gives creators up to four GPUs without the bulk. High-core-count Threadripper PRO keeps heavy scenes responsive. The GPUs drive fast renders, real-time work and creative AI. Every GPU slot runs at full PCIe Gen5 x16 through our MCIO daughterboard and Gen5 retimers, with no riser cables. NVMe drives and high-speed I/O get the lanes they need too. ECC memory support and fast NVMe storage keep large projects stable through long sessions.

CPU icon
1
CPU
RAM (Slots) icon
8
RAM (Slots)
PCIe slots (16x) icon
6
PCIe slots (16x)
PCIe slots (8x) icon
1
PCIe slot (8x)
NVMe M.2 Drives (Gen 5) icon
4
NVMe M.2 Drives (Gen 5)
U.2 Drives icon
2
U.2 Drives
1
Front drive bay
Renderboxes Nano Workstation Internal view showing GPU

Quad GPU
Expansion

Up to four full-length GPUs on our custom PCIe Gen5 MCIO daughterboard. Retimers give every slot full Gen5 x16, with no riser cables.

Renderboxes Nano Workstation Internal view showing airflow

Precision Airflow Engineering

Front and top fan arrays direct airflow for efficient, low-noise cooling under sustained load.

Four full-size GPUs. Every one at full Gen5 bandwidth. Under your desk.

Four full-size GPUs. Every one at full Gen5 bandwidth. Under your desk.
4 full-size GPUs on one board
x16 Gen5 on every GPU slot
384 GB VRAM across four GPUs1
52–54 dBA at 2 m, four GPUs at full load
  1. 1With four NVIDIA RTX PRO 6000 Blackwell GPUs, 96 GB each. Total VRAM depends on the GPUs specified.

Enterprise-grade PCIe connectivity.

The Renderboxes Gen5 daughterboard gives all four GPUs a full PCIe Gen5 x16 link, using server-standard interconnects and signal retiming.

Four full-size GPUs on the Renderboxes Gen5 PCIe daughterboard, each on its own x16 link
Enterprise-standard MCIO interconnect to every GPU.
PCIe Gen5 retimers maintain signal integrity at full x16 bandwidth.
Direct board-to-card connection, with no riser cables in the signal path.
Chassis designed and manufactured in-house, from the GPU out.

Overview

Overview

  • Dimensions (W × H × D): 50.4 cm (54.4 cm with handles) × 57.2 × 47.5 cm
  • Weight: 35kg (65kg in shipping crate)

Technical

  • CPU icon
    AMD Threadripper PRO 9000 Series
  • GPU icon
    NVIDIA RTX (Multi GPU Max 4x)
  • Storage icon
    NVMe Gen 4/5 M.2 / U.2 (Max 2x Hotswap)
  • RAM icon
    DDR5 ECC (Max 2.0TB)
  • Cooling System icon
    Air/AIO

Primary use

The Nano Pro is the world’s only desk-side workstation built to support up to 4 x NVIDIA GeForce RTX 5090 GPUs in a compact tower format. Developed from the foundation of our Atom platform and enhanced for extreme multi-GPU performance, it delivers extraordinary compute density for rendering, AI, simulation, and the most demanding creative workloads.

At the core of the Nano Pro is our groundbreaking GPU decoupling architecture, which physically isolates GPU cooling from the CPU and motherboard in the lower half of the chassis. This allows us to sustain immense GPU power in a footprint that remains practical for studios, edit suites, and under-desk deployment. Compact in size but uncompromising in capability, the Nano Pro is designed for users who need maximum performance without moving to a rackmount environment.

CPU

  • AMD Threadripper PRO 9995WX — 2.5 GHz / up to 5.4 GHz
  • AMD Threadripper PRO 9985WX — 3.2 GHz / up to 5.4 GHz
  • AMD Threadripper PRO 9975WX — 4.0 GHz / up to 5.4 GHz
  • AMD Threadripper PRO 9965WX — 4.2 GHz / up to 5.4 GHz
  • AMD Threadripper PRO 9955WX — 4.5 GHz / up to 5.4 GHz

GPU

  • 4x Full Length PCIe GPU (maximum 4 slot wide, 600 W)

Memory

  • 8x DDR5 ECC RDIMM Slots, max 2.0TB

Storage

  • 2x x4 U.2 Gen 4.0
  • 4x x4 M.2 Gen 5.0

System Cooling

  • 13x Noctua Industrial 120mm
  • 2x Noctua Cromax 80mm
  • 360mm All in one liquid cooler CPU

Connectivity

  • 6x USB 3.2 Gen 2 Type-A ports
  • 2x USB4 Type-C ports
  • 2x 10GbE LAN
  • 1x 1GbE IPMI Management LAN
  • 1x USB-C 20Gbps port
  • VGA port

Dimensions + Weight

  • W × H × D: 50.4 cm (54.4 cm with handles) × 57.2 × 47.5 cm
  • Weight: 35kg (65kg in shipping crate)

Power Supply

  • 2x 2000 watts Platinum ATX

Warranty

  • 3 years on all parts
  • 1 year return to base on all labour
  • 3 years technical support via remote/email/telephone
Front view of the Photon Workstation, with a mesh grill and handle on top.

Photon

  • AMD Ryzen 9000 Series
  • NVIDIA RTX PRO / AMD Radeon AI PRO (Max 1x)
  • NVMe Gen 5/4 M.2 (Max 2x)
  • DDR5 (Max 256GB)
  • Air/AIO
Renderboxes Atom Workstation

Atom

  • AMD Threadripper / Threadripper PRO 9000 Series
  • NVIDIA RTX (Multi GPU Max 4x)
  • NVMe Gen 5 M.2 (Max 3x) / U.2 (Max 2x)
  • DDR5 ECC (Max 2.0TB)
  • Air/AIO
Front view of the Molecule Air Workstation

Molecule Air

  • Dual AMD EPYC 9005 Turin Series
  • NVIDIA RTX PRO (Multi GPU Max 8x)
  • NVMe Gen 4/5 M.2 / U.2 (Max 4x)
  • DDR5 ECC (Max 3.0TB)
  • Air/AIO
Renderboxes Electron Workstation

Electron

  • AMD Ryzen 9000 Series
  • NVIDIA RTX (Max 1x)
  • NVMe Gen 5 M.2 (Max 2x) / U.2 (Max 1x)
  • DDR5 (Max 256GB)
  • Air/AIO
Front view of Renderboxes Nano Pro Workstation

Nano Pro

  • AMD Threadripper PRO 9000 Series
  • NVIDIA RTX (Multi GPU Max 4x)
  • NVMe Gen 4/5 M.2 / U.2 (Max 2x Hotswap)
  • DDR5 ECC (Max 2.0TB)
  • Air/AIO
Front view of Renderboxes Atom Rack Workstation

Atom Rack

  • AMD Threadripper / Threadripper PRO 9000 Series
  • NVIDIA RTX (Multi GPU Max 4x)
  • NVMe Gen 5 M.2 / U.2 (Max 2x)
  • DDR5 ECC (Max 2.0TB)
  • Air/AIO

Nano Pro asks a different question: why should extreme GPU power need extreme space? Four GPUs used to mean server rooms, messy cables and loud fans.
But GPU-heavy work – rendering, real-time engines, AI – happens at desks, not machine rooms. Nano Pro brings that power back to your desk. With four high-end GPUs, it measures 52–54 dBA at 2 m.

Configure

Configure your Nano Pro, or talk to our team about building the right system for the way you work.

What system is right for you?

Renderboxes Workstations

Answer a few simple questions and we’ll recommend a system that is right for your creative needs

Render at the speed of imagination wherever imagination takes you.

Before you buy

As a rule of thumb, a model needs about 2 GB of GPU memory per billion parameters at 16-bit precision, and about half that at 8-bit. With 384 GB of GPU memory1, the Nano Pro holds a 70-billion-parameter model at full precision with room for long conversations and documents, or a far larger model at 8-bit.

The Nano Pro takes four full-power 600 W GPUs or four 300 W efficiency GPUs, both with 96 GB of memory each. Full-power GPUs give the most performance per card; efficiency GPUs draw half the power.

Yes. It measures 52–54 dBA at 2 m with four GPUs at full load, and runs on two 2000 W power supplies, each on its own standard wall socket. No server room or special power supply is needed.

Yes, with RBOS. The Nano Pro can run up to four virtual workstations, each with a dedicated GPU, its own operating system, USB devices and network port. For four workstations we recommend at least 256 GB of system memory, 64 GB per workstation.

Yes. On Linux, the GPUs exchange data peer to peer over their full PCIe Gen5 x16 links, without going through system memory. That is what allows multi-GPU training and large models split across GPUs to scale.

For a team, usually. You pay for one chassis, CPU platform, memory and power system rather than one per user; GPUs are reallocated rather than left idle; and there is one system to secure, update and support.

It measures 50.4 × 57.2 × 47.5 cm (54.4 cm wide with handles) and weighs 35 kg. It ships in a shipping crate at 65 kg, ready to connect.

Up to 2 TB. The configurator goes up to 1 TB; for 2 TB, contact us.

No. Models and data run on the machine, on your own network.

Three years of parts cover, one year of return-to-base labour and three years of technical support.