24.12
Due by December 20, 2024
44% complete
- More RBAC support
- More Relay-compliant GraphQL schema
- Model serving auto-scaling with endpoint metric collection
- New resource slots (hugepages, virtual
cuda.mem) - Prometheus-compatible health/metric of manager and agents
- Per-user UI states & configurations
- More RBAC support
- More Relay-compliant GraphQL schema
- Model serving auto-scaling with endpoint metric collection
- New resource slots (hugepages, virtual
cuda.mem) - Prometheus-compatible health/metric of manager and agents
- Per-user UI states & configurations