Struct GpuInfo 

pub struct GpuInfo {
    pub name: String,
    pub vendor: GpuVendor,
    pub memory_total: usize,
    pub memory_available: usize,
    pub memorybandwidth_gbps: f64,
    pub compute_units: usize,
    pub base_clock_mhz: usize,
    pub memory_clock_mhz: usize,
    pub compute_capability: ComputeCapability,
    pub features: GpuFeatures,
    pub performance: GpuPerformance,
}

GPU information and capabilities

Fields

name: String

GPU name/model

vendor: GpuVendor

GPU vendor

memory_total: usize

Total GPU memory in bytes

memory_available: usize

Available GPU memory in bytes

memorybandwidth_gbps: f64

Memory bandwidth in GB/s

compute_units: usize

Number of compute units (CUDA cores, stream processors, etc.)

base_clock_mhz: usize

Base clock frequency in MHz

memory_clock_mhz: usize

Memory clock frequency in MHz

compute_capability: ComputeCapability

Compute capability/architecture

features: GpuFeatures

Supported features

performance: GpuPerformance

Performance characteristics

Implementations

impl GpuInfo

pub fn detect() -> CoreResult<Self>

Detect GPU information

pub fn performance_score(&self) -> f64

Calculate performance score (0.0 to 1.0)

pub fn optimal_workgroup_size(&self) -> usize

Get optimal workgroup/block size

pub fn is_compute_capable(&self) -> bool

Check if suitable for compute workloads

pub fn is_ml_capable(&self) -> bool

Check if suitable for machine learning

pub fn create_from_pci_ids(vendor_id: &str, device_id: &str) -> Self

Create GpuInfo from PCI IDs

Trait Implementations

impl Clone for GpuInfo

fn clone(&self) -> GpuInfo

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GpuInfo

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

Formats the value using the given formatter.

impl Default for GpuInfo

fn default() -> Self

Returns the “default value” for a type.