#
Model Catalog Statuses
Models move through several statuses during their lifecycle on the PHOENIQS Model Service. This page explains each status and what it means for production use.
#
Experimental models (Preview / Sandbox)
Models with the -preview suffix are experimental. You can use them to explore new capabilities and test application compatibility, but they are not intended for production. Preview models are not covered by the production SLA, have no guaranteed support period, and may be changed or withdrawn without the retirement notice provided for LTS models.
#
Active models
Active models are available for production use. Models designated for Long-Term Support (LTS) come with a defined transition period if we decide to retire them. GLM 5 Flash and DeepSeek 4 Flash are currently designated as LTS. Check the Active Models page for current availability and API identifiers.
#
Scheduled for retirement
When we decide to retire a model, we list it under Scheduled for Retirement. This page provides the planned retirement date and, where available, recommended alternatives and migration guidance. A retirement announcement does not mean that access ends immediately. We may retire a model when a new major version becomes available or when usage remains consistently low.
#
Decommissioned models
The Decommissioned Models page records models that are no longer available in the standard catalog, along with their retirement dates. If your application still depends on one of these models, you can move to an active alternative or contact us about custom hosting.
#
Long-Term Support (LTS)
For an LTS model, PHOENIQS provides support for at least six months after we announce its deprecation. The six months begin on the announcement date, not when the model was first released. LTS protects the major-version API identifier during this transition period. It does not mean that we keep the same underlying minor release unchanged.
A new major version may lead us to retire the previous one, but it is not the only reason. We also review models with consistently low usage. Keeping a model available requires dedicated GPU capacity, even when demand is limited. We balance this cost against customer needs so that we can maintain a sustainable service.