
![]() |
![]() |
![]() |
Data scientists can use the GUI-based SambaStudio platform to train, deploy, and manage models.
|
Developers use the developer stack, which includes predefined models and our Python API.
|
Administrators troubleshoot Runtime, the bridge between hardware and software. Administrators with access to the hardware perform other administration tasks.
|
SambaStudio documentation
-
SambaStudio. Learn how to use SambaStudio. SambaStudio is a rich, GUI-based platform that provides the functionality to train, deploy, and manage models.
-
SambaStudio Introduction. View the introduction to the SambaStudio platform, which provides an overview of the core GUI areas and platform components.
-
SambaStudio CLI Setup. Learn how to install and setup the packages required to access the SambaStudio platform via the SambaNova API (snapi) command-line client.
SambaFlow documentation
-
SambaFlow™ Software Development. Learn how to create models and use existing models with SambaFlow tools and the SambaFlow SDK.
-
SambaFlow Runtime. Learn how to monitor and manage the system service that give access to the SambaNova Reconfigurable Dataflow Unit (RDU) chips.
-
SambaTune. Learn how to run SambaTune and interpret the output. SambaTune supports profiling, debugging and tuning performance of applications running on SambaNova hardware.
DataScale hardware documentation
-
DataScale® Hardware Installation. Learn about site preparation and installation tasks for DataScale hardware.
-
DataScale System Administration. Learn how to manage and troubleshoot the different components of the DataScale rack.