Enum CacheType 

#[non_exhaustive]
pub enum CacheType {
    Ephemeral,
    Extended,
}

Cache type for prompt caching (Anthropic-specific feature)

Controls the time-to-live (TTL) for cached prompt content. Both types offer 90% savings on cache reads, but differ in write costs and duration.

Pricing Model

• Ephemeral writes: 1.25x base input token cost (25% premium)
• Extended writes: 2x base input token cost (100% premium)
• Cache reads (both): 0.1x base input token cost (90% savings)

When to Use

• Ephemeral: Quick iterations, development sessions (< 5 minutes)
• Extended: Long documentation, repeated workflows (< 1 hour)

Example

use multi_llm::{MessageAttributes, CacheType};

// Ephemeral: lower write cost, shorter TTL
let ephemeral = MessageAttributes {
    cacheable: true,
    cache_type: Some(CacheType::Ephemeral),
    ..Default::default()
};

// Extended: higher write cost, longer TTL
let extended = MessageAttributes {
    cacheable: true,
    cache_type: Some(CacheType::Extended),
    ..Default::default()
};

Break-Even Analysis

For 1000 tokens cached and reused N times:

• Ephemeral: Profitable after 1-2 reads (breaks even quickly)
• Extended: Profitable after 5-6 reads (higher initial cost)

See: https://platform.claude.com/docs/en/build-with-claude/prompt-caching

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

Ephemeral

Ephemeral cache (5-minute TTL, 1.25x write cost)

Best for development, quick iterations, and short sessions where you’ll reuse the same context multiple times within 5 minutes.

Extended

Extended cache (1-hour TTL, 2x write cost)

Best for long documentation contexts, extended workflows, or situations where you need the cache to persist across longer time periods (up to 1 hour).

Trait Implementations

impl Clone for CacheType

fn clone(&self) -> CacheType

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for CacheType

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

Formats the value using the given formatter.

impl Default for CacheType

fn default() -> CacheType

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for CacheType

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Hash for CacheType

fn hash<__H: Hasher>(&self, state: &mut __H)

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl PartialEq for CacheType

fn eq(&self, other: &CacheType) -> 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 Serialize for CacheType

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Copy for CacheType

impl Eq for CacheType

impl StructuralPartialEq for CacheType