NBMC-324
(0)
Developer laptop with NVIDIA RTX Spark, 16-inch OLED display, UHD+ resolution, 120 Hz, 128 GB unified memory, 2 TB SSD and Wi-Fi 7
RTX Spark Laptops
With an RTX Spark laptop, you are well equipped for local AI, creative applications and gaming.
3 Results
NBMC-325
(0)
Developer laptop with NVIDIA RTX Spark, 16-inch OLED display, 2.8K resolution, 120 Hz, 64 GB unified memory, 1 TB SSD and Wi-Fi 7
NBMC-326
(0)
Creator laptop with NVIDIA RTX Spark, 16-inch OLED display, 2.8K resolution, 120 Hz, 32 GB unified memory, 1 TB SSD and Wi-Fi 7
RTX Spark Laptops: NVIDIA AI and RTX Graphics in a Single Chip
RTX Spark laptops feature NVIDIA's RTX Spark Superchip, combining AI performance with powerful RTX graphics. Thanks to Unified Memory, these mobile systems are suitable for local AI applications, creative and professional workloads as well as gaming.
What Is RTX Spark?
RTX Spark is an NVIDIA platform for powerful and efficient laptops. At its core is the RTX Spark Superchip, which combines an NVIDIA Grace CPU and a Blackwell RTX GPU with Unified Memory. This shared memory is available to both the CPU and GPU and does not have to be divided into separate memory pools.
Up to 128 GB Unified Memory for Demanding Workloads
One of the key features of RTX Spark is its Unified Memory, which is shared by the CPU and GPU. Depending on the variant, up to 128 GB LPDDR5X is available. The large shared memory pool provides plenty of capacity for memory-intensive applications – from local AI models and creative workflows such as 3D rendering and video editing to professional applications and gaming.
RTX Spark Laptops by Use Case
| Use Case | What to Look For |
|---|---|
| Local AI | Unified Memory capacity, compute performance |
| AI Development | Unified Memory, CUDA/software support |
| Creators | Display, Unified Memory, encoder, ports |
| Gaming | Display, refresh rate, cooling, game compatibility, Unified Memory |
| Mobile Work | Weight, battery capacity, display size |
Local AI with RTX Spark
Whether you are working with local language models, generative AI or developing your own applications, RTX Spark laptops are designed for versatile AI workloads. This allows you to run, test and further develop AI models directly on your device without having to rely on cloud processing for many tasks.
For Creators and Professional Workflows
For video editing, 3D rendering, image editing and complex design projects, RTX Spark provides high compute performance for creative and professional workflows. This allows you to work on extensive projects while on the go and use AI-powered features in compatible applications as part of your workflow.
Memory Requirements for Local AI Models
The amount of memory that needs to be available to the GPU depends on factors such as model size, quantisation and context length. With RTX Spark, the GPU uses part of the Unified Memory shared with the CPU. For models with approximately 4-bit quantisation, the following values can be used as a rough guide:
- 7B to 8B parameters: 5 to 8 GB for the GPU
- 14B parameters: 9 to 12 GB for the GPU
- 32B parameters: 20 to 24 GB for the GPU
- 70B parameters: 40 to 48 GB for the GPU
- 100B+ parameters: depending on the model, more than 50 GB for the GPU
These figures are intended as general guidelines only. Actual memory requirements may vary depending on the model architecture, quantisation, context length, software used and other factors.
RTX Spark for Gaming
In addition to AI applications and professional workflows, RTX Spark laptops can also be used as powerful gaming systems. The integrated RTX graphics based on the Blackwell architecture support modern graphics technologies such as ray tracing to realistically render lighting, shadows and reflections in supported games.
NVIDIA DLSS can further improve image quality and performance in supported games, while NVIDIA Reflex 2 can reduce system latency. RTX Spark supports the RTX platform's current DLSS 5 technology.
Specifications of the RTX Spark N1X Variants
| Feature | RTX Spark N1X | RTX Spark N1X |
|---|---|---|
| GPU | NVIDIA Blackwell RTX GPU with 6,144 cores | NVIDIA Blackwell RTX GPU with 5,120 cores |
| CPU | NVIDIA Grace CPU with 20 cores | NVIDIA Grace CPU with 18 cores |
| Memory | Up to 128 GB LPDDR5X Unified Memory | Up to 128 GB LPDDR5X Unified Memory |
| TDP | 45 – 80 W | |
| Ray Tracing Cores | 4th Generation | |
| Tensor Cores | 5th Generation | |
| NVIDIA DLSS | DLSS 5 | |
| NVIDIA Encoder | 1x 9th Generation | |
| NVIDIA Decoder | 1x 6th Generation | |
| Operating System | Windows 11 | |
How Does RTX Spark Differ from Traditional Laptop Platforms?
RTX Spark combines an Arm-based NVIDIA Grace CPU and a Blackwell RTX GPU in a single system-on-chip. Instead of separate system memory and dedicated graphics memory, the CPU and GPU access Unified Memory. This means that a large shared memory pool can be used for local AI models, creative applications and other memory-intensive workloads, for example.
If you are looking for a compact mini PC specifically for AI, then an NVIDIA DGX Spark Mini PC is the right choice. Thanks to a pre-installed AI software stack, you are equipped for all kinds of AI applications.
FAQ – RTX Spark Laptops
Do RTX Spark Laptops Support CUDA?
Yes. NVIDIA CUDA runs natively on RTX Spark. This allows you to use CUDA-based applications and development environments for AI, computing and other accelerated workloads.
Which NVIDIA Drivers Support RTX Spark Laptops?
RTX Spark supports both NVIDIA Studio Drivers and Game Ready Drivers. This allows you to choose the driver that best suits your focus on creative and professional applications or gaming.
Are RTX Spark Laptops Suitable for AI Development?
Yes. RTX Spark supports the NVIDIA CUDA stack and is designed to develop, test and run AI applications on the same system. This makes the platform suitable for prototyping, inference and working with AI models, for example.
Can I Connect External Monitors to an RTX Spark Laptop?
RTX Spark supports HDMI and DisplayPort. However, the ports actually available on the laptop and the supported resolutions and refresh rates depend on the individual notebook model.
Do Standard Windows Programs and Games Run on RTX Spark Laptops?
RTX Spark runs Windows 11 on an Arm-based platform. Arm64 applications can run natively, while many x86 and x64 programs run via Microsoft's Prism emulation. However, compatibility may vary depending on the application, game, driver or additional software. For specialist software, you should therefore check Windows on Arm support before purchasing.
How Long Does the Battery of an RTX Spark Laptop Last?
Battery life varies depending on the laptop, battery capacity and usage. NVIDIA has designed RTX Spark with high energy efficiency in mind and advertises corresponding laptops with all-day battery life. You can find the specific battery life in the specifications provided by the respective manufacturer.