Struct ModelSpec 

pub struct ModelSpec {
    pub prefix: &'static str,
    pub context_window: u64,
    pub compaction_buffer_pct: f64,
}

Model specification with named fields for type safety

NOTE: Why struct instead of tuple (&str, u64)? Tuples are position-dependent and lack semantic meaning:

• (“claude-3-5”, 200_000) vs (200_000, “claude-3-5”) - compiler can’t catch order mistakes
• No field names make code less self-documenting
• Hard to extend (adding output_limit would create complex tuples)

Structs provide:

• Named fields prevent position errors (can’t swap prefix and context_window)
• Self-documenting code (field names explain purpose)
• Easy to extend with new fields (e.g., output_limit, cache_support)
• IDE auto-completion works
• Compiler enforces all fields are provided

Fields

prefix: &'static str

context_window: u64

compaction_buffer_pct: f64

Compaction buffer percentage (0-100) When input tokens exceed (100% - compaction_buffer_pct), compaction is triggered

Implementations

impl ModelSpec

pub const fn new( prefix: &'static str, context_window: u64, compaction_buffer_pct: f64, ) -> Self

Create a new model specification

NOTE: Why const fn? Allows construction at compile time in const context (MODEL_SPECS array). Zero runtime overhead - all values computed at compile time.

Trait Implementations

impl Clone for ModelSpec

fn clone(&self) -> ModelSpec

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ModelSpec

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

Formats the value using the given formatter.

impl PartialEq for ModelSpec

fn eq(&self, other: &ModelSpec) -> 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 ModelSpec

impl StructuralPartialEq for ModelSpec