Online
Qwen3-14B
Primary inference model
127.0.0.1:8081
GGUF · Q4_K_M · llama.cpp
Private infrastructure
System overview
All core services operational
VS
Enterprise self-hosted AI platform
One control plane for models, inference, observability and private AI services.
VentusIS AI Admin is the internal operations center for managing the local LLM stack. It is designed to grow into model lifecycle management, API governance, RAG administration, usage analytics and secure access control.
Online
Prometheus
Metrics collection
127.0.0.1:9090
System + inference metrics
Online
Grafana
Observability dashboards
127.0.0.1:3001
Private operations access
Platform architecture
Current production topology
Public AI Gatewayai.ventusis.comPlanned
Admin Control Planeadmin.ai.ventusis.comActive
Inference Runtimellama.cpp + Qwen3-14BActive
ObservabilityPrometheus + GrafanaActive
Roadmap
Next production capabilities
01Authentication and RBAC
02Live metrics from Prometheus
03Model registry and lifecycle
04API keys, limits and audit log
05RAG knowledge administration