The evolution of AI in a nutshell – NVIDIA DGX Spark reaches developers all over the world!

Calendar 10/15/2025

DGX Spark – a desktop AI supercomputer with 1 PFLOP of power and 128 GB of memory. Accelerate model and agent development locally without the cloud!

NVIDIA has just begun the global shipment of DGX Spark – the smallest AI supercomputer in the company's history. The system is designed to provide artificial intelligence creators with local server-class computing power, packaged in a desktop format. In practice, this means 1 PFLOP of computing power and 128 GB of unified memory, enabling the running of models with up to 200 billion parameters without the need to utilise cloud services.

The launch was not without symbolism – Jensen Huang, CEO of NVIDIA, personally delivered the first unit to Elon Musk at SpaceX, referencing 2016 when the same duo collected the first DGX-1 system, which later gave rise to ChatGPT.

“In 2016, DGX-1 gave researchers their own supercomputer. Now DGX Spark is set to put AI in the hands of every developer” – said Jensen Huang.

AI power in a desktop format

The new DGX Spark is based on the Grace Blackwell architecture and is equipped with the GB10 Superchip, which combines GPU and CPU with a bandwidth 5x greater than PCIe 5.0. The entire system consumes only 240 W of power, weighs 1.2 kg, yet offers petascale data processing.

Included is the complete NVIDIA AI Stack environment – CUDA, NVLink-C2C, ConnectX-7 (200 Gb/s), libraries, models, and microservices NVIDIA NIM, allowing AI projects to be launched immediately after unboxing the hardware.

The first units are going to partners such as Acer, ASUS, Dell, Lenovo, MSI, HP, GIGABYTE, as well as Anaconda, Meta, Microsoft, Hugging Face, and JetBrains. NVIDIA announces that DGX Spark will become the new standard for creating AI agents and physical models, including in local applications.

Comparison: from DGX-1 to DGX Spark

Specification

DGX-1 (2016)

DGX Spark (2025)

GPU Architecture

NVIDIA Pascal

NVIDIA Blackwell

GPU Memory

128 GB (16 GB per GPU)

128 GB unified system memory

AI Performance

170 TFLOPS (FP16)

1 PFLOP (FP4)

Power Consumption

3200 W

240 W

Dimensions

866 × 444 × 131 mm

150 × 150 × 50.5 mm

Weight

60.8 kg

1.2 kg

Price

129,000 USD

3,999 USD

A New Era for AI Creators

With Spark, developers can locally train and fine-tune models with up to 70 billion parameters, create AI agents, tools for computer vision, or personalised chatbots optimised for Grace Blackwell. The system is fully prepared to work with models such as FLUX.1, Qwen3 or Cosmos Reason.

As Professor Kyunghyun Cho from NYU Global AI Frontier Lab put it:

“DGX Spark gives us petascale power on the desktop. It’s a completely new way of conducting AI research — even in sensitive areas such as medicine.”

From 15th October DGX Spark can be ordered at NVIDIA.com, and partners such as Acer, ASUS, Dell, HP, Lenovo and MSI will offer their own versions of this computer. It will also be available in the Micro Center network in the USA, and global distribution will commence in the coming weeks.

Katarzyna Petru Avatar
Katarzyna Petru

Journalist, reviewer, and columnist for the "ChooseTV" portal