Struct ModelEndpoint 

pub struct ModelEndpoint {
    pub id: ModelId,
    pub display_name: String,
    pub description: String,
    pub base_model: String,
    pub adapter: Option<String>,
    pub max_context_length: usize,
    pub is_default: bool,
    pub status: EndpointStatus,
}

Metadata for a served model variant.

Each endpoint represents a unique model configuration that can receive inference requests. A base model may have multiple endpoints, each with a different LoRA adapter applied.

Fields

id: ModelId

Unique identifier for this endpoint.

display_name: String

Human-readable display name.

description: String

Longer description of what this endpoint provides.

base_model: String

Name of the underlying base model.

adapter: Option<String>

Optional LoRA adapter name applied on top of the base model.

max_context_length: usize

Maximum context length (in tokens) this endpoint supports.

is_default: bool

Whether this endpoint is the default when no model is specified.

status: EndpointStatus

Current operational status.

Implementations

impl ModelEndpoint

pub fn new(id: impl Into<String>, base_model: impl Into<String>) -> Self

Create a new endpoint with sensible defaults.

Status is set to Ready, no adapter, default context length of 4096.

pub fn with_adapter(self, adapter: impl Into<String>) -> Self

Attach a LoRA adapter to this endpoint.

pub fn with_description(self, desc: impl Into<String>) -> Self

Set a human-readable description.

pub fn with_context_length(self, ctx: usize) -> Self

Set the maximum context length.

pub fn set_default(self) -> Self

Mark this endpoint as the default.

Trait Implementations

impl Clone for ModelEndpoint

fn clone(&self) -> ModelEndpoint

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ModelEndpoint

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

Formats the value using the given formatter.