Running a 120B model offline on Surface RTX Spark Dev Box with Intelligent Terminal
Microsoft Developer shows how Intelligent Terminal uses a locally hosted model for context-aware development tasks on Surface hardware, with inference running on the device rather than in the cloud.
Local model execution
The demonstration features a 120-billion-parameter model running entirely locally on an NVIDIA RTX Spark N1X with 128 GB of unified memory. The video title highlights Surface RTX Spark Dev Box, while the description identifies Surface Laptop Ultra as the laptop used for local execution.
Intelligent Terminal uses the local model to complete complex workflows, including writing new agent skills with context, without sending requests to the cloud.