Cloud World Model on Smithery
    Cloud World Model
    Reference
    Updated August 25, 2026

    Provider Coverage

    Every service, instance family, and pricing model the simulator models for each cloud provider. Counts and base rates are derived from the canonical pricing constants — the same numbers used by the simulation engine.

    Accuracy label guide
    Measured
    Validated against real provider benchmarks — latency, cost, and CPU metrics are fit to reference data from the provider's own documentation and publicly available performance studies.
    Extrapolated
    Derived from the simulation model — behavior is modeled from first principles or scaled from adjacent measured services, without a direct benchmark validation pass. Suitable for architectural comparisons; use real infrastructure for production SLO validation.

    AWS

    aws
    Compute sizes
    244
    instance types
    DB sizes
    16
    managed DB tiers
    Base latency
    45 ms
    intra-region baseline
    Compute rate
    $0.096/hr
    reference instance
    Modeled Services
    Compute
    EC2 (t2, t3, m5, m6i, m7a, m7g, m7gd, m7i, m7i-flex, m8a, m8azn, m8g, m8gb, m8gd, m8gn, m8i, m8i-flex, m8ib, m8id, m8idb, m8idn, m8in, m8ine, m9g, m9gd, c5, c6i, r5, r6i)
    Measured
    Network
    Application Load Balancer (ALB)
    Measured
    CloudFront CDN
    Extrapolated
    NAT Gateway
    hourly + GB
    Extrapolated
    VPC Endpoint (Interface)
    hourly
    Extrapolated
    Elastic IP (unattached)
    hourly
    Extrapolated
    Database
    RDS (PostgreSQL / MySQL, Graviton3 families)
    Measured
    Aurora PostgreSQL
    Measured
    Aurora DSQL
    request-based
    Extrapolated
    Aurora Serverless v2
    ACU-hour
    Extrapolated
    DynamoDB
    request-based
    Extrapolated
    Cache
    ElastiCache (Redis)
    Measured
    Storage
    S3
    GB-month
    Extrapolated
    EBS (sc1, st1, gp3, io1, io2 Block Express)
    Measured
    Serverless
    Lambda
    GB-second
    Extrapolated
    ECS Fargate (Linux task fleets)
    vCPU-second + GB-second
    Extrapolated
    App Runner
    vCPU-second
    Extrapolated
    Queue / Streaming
    Amazon SQS
    request-based
    Extrapolated
    Kubernetes
    Amazon EKS
    Extrapolated
    ROSA Hosted Control Planes (AWS) — published $0.171/4-vCPU worker service fee + $0.25 cluster fee
    platform + control plane + AWS workers
    Extrapolated
    ROSA Classic (AWS) — published $0.171/4-vCPU worker service fee
    platform + AWS workers
    Extrapolated
    OpenShift Dedicated (AWS) — OpenShift platform overlay; subscription pricing estimated
    platform fee + AWS workers
    Extrapolated
    Self-managed OpenShift (AWS) — subscription pricing estimated
    AWS workers + estimated subscription
    Extrapolated
    EKS GPU node group (g5 / NVIDIA A10G)
    Measured
    EKS large-GPU node group (p4d.24xlarge 8×A100, p3.16xlarge 8×V100, g5.48xlarge 8×A10G)
    Extrapolated
    Observability
    CloudWatch
    GB-ingested
    Extrapolated

    Compute size count (244) is derived from the live size array in shared/provider-sizes.ts. Base compute rate ($0.096/hr) and base latency (45 ms) come from shared/provider-pricing.ts and shared/provider-config.ts respectively.

    Coverage snapshot version: 2026-08-25. This snapshot is validated against a 90-day freshness window in CI. To update, bump COVERAGE_VERSION in shared/provider-coverage.ts after reviewing pricing constants.