truffle-ts API reference - v0.4.2
    Preparing search index...

    Interface InstanceType

    One EC2 instance type's capabilities. This is the shape of a bundled-catalog entry and of a Finder.search result. Optional fields are omitted when zero/ unknown, matching the Go omitempty JSON tags. Prices are approximate static estimates ("as of 2026-01"), never live.

    interface InstanceType {
        architecture: Architecture;
        estimatedPrice?: boolean;
        gpuManufacturer?: string;
        gpuMemoryMib?: number;
        gpuModel?: string;
        gpus?: number;
        instanceFamily: string;
        instanceType: string;
        memoryMib: number;
        nestedVirt?: boolean;
        onDemandPrice?: number;
        physicalCores?: number;
        threadsPerCore?: number;
        vcpus: number;
    }
    Index
    architecture: Architecture
    estimatedPrice?: boolean

    True when this entry's specs/price are NOT live AWS data — carried over from the hand seed because the type isn't offered in the generated region (legacy families like g3/p2/p3, or brand-new ones). Absent = came from the live pull.

    gpuManufacturer?: string

    GPU vendor, e.g. "nvidia".

    gpuMemoryMib?: number

    Total GPU memory in MiB across all GPUs.

    gpuModel?: string

    GPU model name, e.g. "A100".

    gpus?: number

    Number of GPUs; omitted for non-GPU instances.

    instanceFamily: string

    Family prefix, e.g. "m6i".

    instanceType: string

    e.g. "m6i.2xlarge".

    memoryMib: number

    Memory in MiB.

    nestedVirt?: boolean

    True if the type supports nested virtualization (KVM/Hyper-V in-instance).

    onDemandPrice?: number

    On-demand $/hr. Live from the Pricing API when regenerated, else a static estimate.

    physicalCores?: number

    Physical CPU cores (vCPUs / threadsPerCore).

    threadsPerCore?: number

    Threads per physical core (1 for Graviton, 2 for most x86).

    vcpus: number