Understand this story
The New Stack conducted hands-on testing of local AI model performance on Microsoft's Surface Laptop Ultra. The tested model uses the Nvidia N1X platform, a Blackwell RTX GPU, and 64GB of memory, with a configured price of $4,300. Test results showed that Nemotron 3.5 Bolt reached 56-60 tokens/s; 8-bit Qwen 3.8 27B managed only 8-9 tokens/s, and rose to 25-28 tokens/s after dropping to 4-bit. After unplugging, AI workload performance dropped by about 20%-25%, which does not match Microsoft's claim of retaining 99.6%. There are currently no further developments.
AI-generated from public reports
Latest development · Report
Running local AI models on the Microsoft Surface Laptop Ultra: hands-on findingsDevelopments
1 confirmed developments
The New Stack tests local AI workloads on the Surface Laptop Ultra
Source:The New Stack
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 16
The New Stack
Running local AI models on the Microsoft Surface Laptop Ultra: hands-on findings
The New Stack tested running local AI models on the Microsoft Surface Laptop Ultra (Nvidia N1X platform, Blackwell RTX GPU, 64GB of memory, $4,300 configuration): Nemotron 3.5 Bolt reached 56-60 tokens/s, while 8-bit Qwen 3.8 27B managed only 8-9 tokens/s, recovering to 25-28 tokens/s after dropping to 4-bit. On battery, AI workload performance dropped by about 20%-25%, inconsistent with Microsoft's claim of retaining 99.6%.