Enum BenchRequestSourceInput 

pub enum BenchRequestSourceInput {
    Random {
        input_tokens: BenchTokenSelectorInput,
        output_tokens: BenchTokenSelectorInput,
        prefix_sharing: Option<BenchPrefixSharingInput>,
        shared_system_content: Option<BenchSharedSystemContentInput>,
        corpus: Option<BenchCorpusInput>,
        images: Option<BenchImagesInput>,
    },
    RandomMixture {
        shapes: Vec<BenchRandomShapeInput>,
        total_weight: u64,
        prefix_sharing: Option<BenchPrefixSharingInput>,
    },
    Dataset {
        dataset: String,
        profile: Option<String>,
        max_input_tokens: u32,
        output_tokens: Option<u32>,
        catalog: Box<BenchDatasetCatalogInput>,
    },
    Replay {
        path: String,
        expected_sha256: Option<String>,
        prefix_sharing: Option<BenchPrefixSharingInput>,
    },
}

One closed request origin lowered by Inferlab for the Bench runtime.

Variants

Random

AIPerf generates exact token-shape prompts from the release-pinned synthetic generator.

Fields

input_tokens: BenchTokenSelectorInput

output_tokens: BenchTokenSelectorInput

prefix_sharing: Option<BenchPrefixSharingInput>

shared_system_content: Option<BenchSharedSystemContentInput>

corpus: Option<BenchCorpusInput>

images: Option<BenchImagesInput>

The effective per-request image decoration ([[RFC-0004:C-BENCH-REQUEST-SOURCES]]); absent when the source declares no images.

RandomMixture

AIPerf samples exact token-shape pairs from one seeded categorical distribution.

Fields

shapes: Vec<BenchRandomShapeInput>

total_weight: u64

prefix_sharing: Option<BenchPrefixSharingInput>

Dataset

Inferlab materializes a release-catalog conversation snapshot before AIPerf starts.

Fields

dataset: String

profile: Option<String>

max_input_tokens: u32

output_tokens: Option<u32>

catalog: Box<BenchDatasetCatalogInput>

Replay

Inferlab replays one workspace-local frozen population file unchanged; the file bytes are the sole population authority ([[RFC-0004:C-BENCH-REQUEST-SOURCES]]).

Fields

path: String

Workspace-relative population path as declared. The preparation request’s source_path carries its absolute resolution.

expected_sha256: Option<String>

prefix_sharing: Option<BenchPrefixSharingInput>

Trait Implementations

impl Clone for BenchRequestSourceInput

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for BenchRequestSourceInput

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for BenchRequestSourceInput

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl JsonSchema for BenchRequestSourceInput

fn schema_name() -> Cow<'static, str>

The name of the generated JSON Schema.

fn schema_id() -> Cow<'static, str>

Returns a string that uniquely identifies the schema produced by this type.

fn json_schema(generator: &mut SchemaGenerator) -> Schema

Generates a JSON Schema for this type.

fn inline_schema() -> bool

Whether JSON Schemas generated for this type should be included directly in parent schemas, rather than being re-used where possible using the $ref keyword.

impl PartialEq for BenchRequestSourceInput

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

Equality operator ==.

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

Inequality operator !=.

impl Serialize for BenchRequestSourceInput

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 BenchRequestSourceInput