Parallel AI stack for Omics Archive
BioNeMo + vLLM + NIM‑compatible inference in a parallel environment.
Checking auth status...
Per‑user limits and remaining quota.
Activity summary and recent job volume.
Upload a model artifact to GCS, then activate it for vLLM.
Manifests discovered from GCS.
Model roots in the models bucket.
Parsed model_card.json from GCS.
Configure TTL and dry‑run flag in Kubernetes secret.
Recent cleanup configuration changes.
/api/health — API health
/v1/health/ready — NIM‑like health
/v1/models — NIM‑like model list