Processor
- Manufacturer
- Nvidia
- Family
- -
- Model
- GB10
- Frequency
- 3.5 GHz
RAM Memory
- Type
- DDR5
- Size
- 128 GB
Hard Drive
- Type
- SSD
- Capacity
- 2000 GB
Graphics Card
- Manufacturer
- nVidia
Product Details
- Suggested Use
- Performance
- Keyboard Language
- No Keyboard
- RGB Lighting
- No
- Colour
- Black
- Popular Series
- Workstation
Software
- Operating System
- No Operating System
Top specs
- Processor
- GB10
- GPU
- nVidia
Important information
Specifications are collected from official manufacturer websites. Please verify the specifications before proceeding with your final purchase. If you notice any problem you can report it here.
Description
The Lenovo ThinkStation PGX is the first workstation dedicated exclusively to developing and deploying artificial intelligence applications. This compact AI workstation is built around the revolutionary NVIDIA GB10 Grace Blackwell Superchip.
It features NVIDIA DGX OS and the full NVIDIA AI software stack, including tools such as PyTorch and Jupyter® Notebooks. Engineers can easily prototype, fine-tune and run inference on models, ensuring smooth transfer and deployment to data center or cloud platforms.
Thanks to its compact size, energy efficiency and data security, it is suitable for labs, startups or businesses looking for AI power on-site. The NVIDIA GB10 Grace Blackwell Superchip combines a multicore ARM CPU (20 cores) with cutting-edge integrated graphics, delivering an energy-efficient System-on-a-Chip (SoC) solution.
With 1 PetaFLOP of AI computing power and support for models with up to 200 billion parameters (and up to 405 billion with two PGX systems), the PGX delivers speed and reliability. The ConnectX®-7 Smart NIC (10GbE) network port lets you connect two ThinkStation PGX systems (an additional cable is required) to double the computing power and handle larger AI models.
The 128GB of LPDDR5x memory enables seamless work with large AI models locally. It provides a secure “sandbox” environment for prototyping AI models, keeping your intellectual property (IP) on-premises and reducing security risks.
Ideal for developers, researchers and data scientists who want to develop, fine-tune or run large AI models without relying on the cloud.