AI has moved into everyday work. Developers pair with coding assistants, editors clean up audio in a click, and analysts ask questions of data that used to need a specialist. But a lot of that work still stalls in the same three places. Big jobs queue for a workstation or cloud capacity. Every prompt runs up a usage bill. And some data simply isn’t allowed to leave the building.
Surface Laptop Ultra for Business takes on all three. Co-engineered with NVIDIA and built around NVIDIA RTX Spark™, it handles the work that usually needs a desktop workstation: compiling code, running simulations, rendering 3D and building AI models. Then it goes home in your bag.
It’s not standard issue. Here’s what it changes about the working day.
AI that keeps up with your thinking
The headline number is up to 1 petaflop of AI compute. In practice, that means you can run AI models of up to 120 billion parameters locally, on the laptop in front of you, with up to 128GB of unified memory for the model and its data to draw on.
For a developer, AI-assisted code completion runs on the device, so suggestions keep pace with your typing. For a video editor, AI noise reduction, intelligent masking and upscaling happen in the edit rather than in a render queue. For a data scientist, models and datasets that used to live on a server can be explored wherever the work happens to be. With full support for NVIDIA CUDA alongside the Windows AI stack, the tools your team already relies on come along too.
Big jobs finish sooner, and the thinking doesn’t stop while you wait for them.
Token anxiety, meet your match
Cloud AI is billed in tokens, the small chunks of text and data a model reads and writes. Every prompt, every test run and every ‘let’s try that again’ adds to the meter. For teams building and testing AI, that creates a quiet hesitation: is this experiment worth what it costs?
When AI runs on your own device, that hesitation goes. Developers can test an agent again and again without watching a cloud usage meter tick over. Designers can generate one more variation. Teams can experiment freely, then save the cloud for the work that needs its scale.
That’s the hybrid approach Microsoft is building towards: local AI for speed, privacy and working offline, and the cloud for scale and collaboration. Your cloud costs become something you manage, rather than something that happens to you.
Sensitive work stays local
Some of the most valuable AI work involves the data you’re least comfortable sending anywhere: client records, unreleased designs, source code and financials. Surface Laptop Ultra lets your team process that work directly on the device, with greater control and lower latency than cloud-based alternatives.
Protecting it is hardware-rooted security that guards credentials, keys and firmware wherever the laptop goes. Underneath, a memory-safe architecture combines Rust-based firmware, a Secure Embedded Controller, a discrete TPM and Microsoft Pluton to protect the device from boot through to runtime. It’s a Windows 11 Secured-core PC, Windows Hello signs you in with your face instead of a password, and the user-removable SSD means storage can stay in-house when a device is retired or repaired.
For IT, that’s a rare combination: more power for users and more control for the business. And because it’s Surface, it deploys and is managed through Windows Autopilot and Intune like the rest of your fleet.
Desktop power with laptop freedom
Power only matters if it lasts the length of a real job. An all-new thermal system gives Surface Laptop Ultra up to 2.5 times the thermal capacity of Surface Laptop 15-inch (7th Edition), designed for sustained high performance. It’s also built to hold that performance on battery, so the long compile or the final export doesn’t have to wait for a power point.
The rest of the design is about removing small frictions. The 15-inch mini-LED PixelSense Ultra display, with 262 ppi and up to 2,000 nits of peak HDR brightness, is made for people making critical colour and exposure decisions. USB-C, USB-A, HDMI, a full-size SD card reader and a headphone jack mean no dongles. Back at the desk, it drives up to three 4K external displays. The whole thing is under 18mm thick and under 2 kg, and the charger fits in a jacket pocket.
Where developers, engineers and creators do their best work
Surface Laptop Ultra isn’t meant for every desk. It’s built for the roles where waiting on hardware, cloud capacity or a data approval costs real time. For many others, Surface Laptop or Surface Pro will do the job well at a lower cost. The skill is knowing who sits where.
That’s the conversation we like having. ASI has spent more than 40 years helping Australian organisations put the right technology in the right hands, and as a Microsoft Solutions Partner we know the Surface range well. We’ll help you work out which roles will feel the difference, then look after configuration, deployment and support so devices arrive ready to work.
Surface Laptop Ultra for Business is arriving in Australia soon. Find out how it could help your team do more of its best work.
Talk to our team about what the new Surface for Business devices could mean for your fleet.
