Struct DeviceCaps 

pub struct DeviceCaps {
    pub name: String,
    pub backend: String,
    pub device_type: String,
    pub max_storage_buffer_binding_size: u64,
    pub max_buffer_size: u64,
    pub max_compute_workgroup_size_x: u32,
    pub max_compute_invocations_per_workgroup: u32,
    pub features: String,
}

Hardware capabilities consumed by the application-layer device planner (sharding, KV budget, prefill chunk sizing). Serialized to JSON across the FFI boundary.

Fields

name: String

Human-readable adapter name (e.g. “Apple M3 Pro”).

backend: String

Graphics backend in use (“Metal”, “Vulkan”, “Dx12”, “Gl”, …).

device_type: String

Device type (“IntegratedGpu”, “DiscreteGpu”, …).

max_storage_buffer_binding_size: u64

max_storage_buffer_binding_size: the hard cap on a single GPU buffer — the sharding limit on mobile devices.

max_buffer_size: u64

max_buffer_size: the largest single allocation the driver allows.

max_compute_workgroup_size_x: u32

Largest compute workgroup dimension.

max_compute_invocations_per_workgroup: u32

max_compute_invocations_per_workgroup.

features: String

Debug dump of the adapter’s enabled feature set (wgpu 29 no longer exposes WebGPU extension features like SHADER_F16 through the public adapter API, so we surface the raw list for the planner).

Trait Implementations

impl Clone for DeviceCaps

fn clone(&self) -> DeviceCaps

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for DeviceCaps

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Serialize for DeviceCaps

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.