Enum GpuClass 

pub enum GpuClass {
    A10,
    A100,
    H100,
    Rtx4090,
    Rtx3090,
    Unknown,
}

GPU class with expected throughput ranges.

From cbtop spec §21.7 “Expected Throughput by GPU Class”.

Variants

A10

NVIDIA A10 (24GB) - Data center inference GPU

A100

NVIDIA A100 (40GB or 80GB) - Data center training/inference GPU

H100

NVIDIA H100 (80GB) - Hopper architecture flagship

Rtx4090

NVIDIA RTX 4090 (24GB) - Consumer flagship

Rtx3090

NVIDIA RTX 3090 (24GB) - Previous gen consumer flagship

Unknown

Unknown GPU class

Implementations

impl GpuClass

pub fn expected_throughput(&self) -> (u32, u32)

Expected throughput range (min, max) in tok/s.

From cbtop spec §21.7.

pub fn vram_gb(&self) -> u32

VRAM size in GB.

pub fn from_name(name: &str) -> Self

Detect GPU class from GPU name string.

Parses common GPU name formats from nvidia-smi, NVML, etc.

Trait Implementations

impl Clone for GpuClass

fn clone(&self) -> GpuClass

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GpuClass

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

Formats the value using the given formatter.

impl Display for GpuClass

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

Formats the value using the given formatter.

impl PartialEq for GpuClass

fn eq(&self, other: &GpuClass) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Copy for GpuClass

impl Eq for GpuClass

impl StructuralPartialEq for GpuClass