Cloud platform · Product scope
Small Language Model Hosting
SLM means small language model. Confirm supported model formats and runtimes, hardware requirements, GPU availability, deployment and monitoring, data flow, and whether inference is local or uses an external provider.
Status: Availability and support scope to confirm. This page lists evaluation topics, not verified product features.
Questions to settle before deployment
Which models and runtimes are required? What hardware and performance limits apply? Where does inference run, and what data reaches any provider?
GPU availability, private inference, model support, and data-retention behavior are not promised here.
Agree responsibilities and limitations
Before work begins, document prerequisites, customer actions, UniSmart’s service boundary, operational hours, support route, licensing, and any recovery expectations that apply.