This screen manages training results as versions of named models. Each version carries a stage (whether it is being validated or in operation), and moving a version to operation (Production) always goes through approval. The container image used for serving and field deployment is also built from the version screen.

Model list

Model registry — ① filter by Production ② search ③ versions per stage ④ a model (opens detail)
#AreaDescription
Production filterAll models (전체 모델) · Has Production (Production 있음) · No Production (Production 없음).
Search models (모델 검색)Find by model name.
Stage columnsFor each stage, the version in that stage. "—" if none.
ModelClick to open the model detail. Models marked as safety-critical get a Safety-critical (안전 필수) badge.

Model detail

Model detail — ① drift, monitoring, delete model ② promote to Production ③ version history ④ archive ⑤ compare versions ⑥ comments
#AreaDescription
Header actionsShortcuts to Drift (드리프트) and Monitoring (모니터링), Delete model (모델 삭제) (model_register). Deletion is blocked if there is a Production version.
Promote to production (프로덕션 승격)Promotion request dialog (production_promote, APPROVER, ADMIN).
Version history (버전 히스토리)Version, stage, status, source run, created. Click a row to open the version detail.
ArchiveMoves the version down to Archived (staging_promote; production_promote for a Production version). The trash icon on the right deletes the version (model_register).
Compare versions (버전 비교)Pick two versions to see their run metrics side by side.
Comments (코멘트)Notes per version. Writing needs develop.

If the approvals feature is on and there is a pending promotion request, a link such as "Awaiting approval: v3 (0/1) (승인 대기 중: v3 (0/1))" appears under the title.

Model version detail

Model version — ① version info (download, archive, delete) ② container image (serving image build) ③ comments
#AreaDescription
Version info (버전 정보)Version, stage, status, run, created. Download model (모델 다운로드), Archive (아카이브), Delete version (버전 삭제).
Container image (컨테이너 이미지)The serving image of this version. Create it with Build image (이미지 빌드) (deploy); if one already exists, the button becomes Rebuild (다시 빌드). The built image is reused by API services and field deployment.
Comments (코멘트)Same as the comments on the model detail.

If building is not possible in this environment, a yellow notice appears in ② (for example "There is no serving builder variant (서빙 빌더 변형이 없습니다)" — the tenant administrator must publish a serving builder). During a build, watch the build log live with View logs (로그 보기); if it fails, check the cause with Build logs (빌드 로그).

Main tasks

Request Production promotion

  1. On the model detail, press Promote to production (프로덕션 승격).
  2. Choose the version to promote (승격 버전) and enter a reason (사유) (required). Rollback conditions (롤백 조건) and scope (범위) are recorded as grounds for the approver's judgment. Press Request promotion (승격 요청).
    Request Production promotion — it goes to Approvals and becomes Production once approved
  3. The request goes to the Approvals screen. Once the required approvals are collected, that version becomes Production and the previous Production version moves down to Archived.

Even after promotion, deployment does not start automatically. Serving and field deployment are done separately in API services and Deployments.

Archive

Click the archive icon on a version row and confirm to move that version down to Archived. Only the registry label changes; nothing is deleted.

Archive a version — only the registry label changes; nothing is deleted

Statuses and badges

StageMeaningHow a version gets there
NoneNot assigned
StagingBeing validatedVersions go here when you register a model from a run or training registers it automatically.
ProductionIn operationPromotion request → approval. One per model.
ArchivedKept for referenceArchived by hand, or pushed out when a new version becomes Production.

Image build status: Building (빌드 중) (BUILDING) · Ready (준비됨) (READY) · Failed (실패) (FAILED).

Written for the platform as of 2026-09-21.

© Geo-MLOps