Struct ModelResourceProfile 

pub struct ModelResourceProfile {

    pub schema_version: u32,
    pub path: PathBuf,
    pub artifact_format: ArtifactFormat,
    pub model_family: Option<String>,
    pub architecture: Option<String>,
    pub tensor_count: Option<usize>,
    pub checkpoint_shards: Option<usize>,
    pub embedded_draft_layers: Observed<usize>,
    pub stored_tensor_bytes: Observed<u64>,
    pub largest_stored_tensor_bytes: Observed<u64>,
    pub materialized_parameter_bytes: Observed<u64>,
    pub pinned_parameter_bytes: Observed<u64>,
    pub largest_execution_group_bytes: Observed<u64>,
    pub largest_adjacent_execution_groups_bytes: Observed<u64>,
    pub expert_parameter_bytes: Observed<u64>,
}

Architecture and header-derived planning facts used before a model is loaded.

Fields

schema_version: u32

Version of this serialized resource schema.

path: PathBuf

Inspected checkpoint path.

artifact_format: ArtifactFormat

Physical checkpoint container.

model_family: Option<String>

Resolved model family, when architecture inspection succeeded.

architecture: Option<String>

Resolved architecture name, when available.

tensor_count: Option<usize>

Number of logical tensors exposed by the checkpoint catalog.

checkpoint_shards: Option<usize>

Number of physical checkpoint shards.

embedded_draft_layers: Observed<usize>

Embedded prediction depth derived from normalized architecture policy.

stored_tensor_bytes: Observed<u64>

Sum of encoded tensor payload bytes, excluding container metadata.

largest_stored_tensor_bytes: Observed<u64>

Largest encoded logical or physical tensor payload.

materialized_parameter_bytes: Observed<u64>

Expected execution-time parameter bytes after translation or quantization.

pinned_parameter_bytes: Observed<u64>

Bytes in parameters pinned outside repeated execution groups.

largest_execution_group_bytes: Observed<u64>

Largest single repeated execution group.

largest_adjacent_execution_groups_bytes: Observed<u64>

Largest adjacent pair required by dense streaming’s device window.

expert_parameter_bytes: Observed<u64>

Total routed-expert bytes, where the architecture exposes an exact plan.

Implementations

impl ModelResourceProfile

pub fn unmeasured(path: PathBuf, artifact_format: ArtifactFormat) -> Self

Creates an explicitly unmeasured resource profile.

Trait Implementations

impl Clone for ModelResourceProfile

fn clone(&self) -> ModelResourceProfile

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ModelResourceProfile

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ModelResourceProfile

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for ModelResourceProfile

impl PartialEq for ModelResourceProfile

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

Equality operator ==.

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

Inequality operator !=.

impl Serialize for ModelResourceProfile

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for ModelResourceProfile