Function calculate_cost 

pub fn calculate_cost(
    model: &str,
    input_tokens: u64,
    output_tokens: u64,
    cache_read: u64,
    cache_creation: u64,
) -> f64

Calculate the USD cost of a single model turn (no cache-write TTL split).

Thin wrapper over calculate_cost_split: the aggregate cache_creation count is billed at the 5m write rate (1.25×). When the user opted into 1h caching but the TTL split didn’t arrive, this under-bills (fail-cheap, not fail-expensive — cost display is informational, not invoiced).

Arguments

• model – Model identifier string (e.g. "claude-sonnet-4-5").
• input_tokens – Uncached input tokens billed at full input rate.
• output_tokens – Output / generated tokens (includes adaptive thinking).
• cache_read – Tokens served from the prompt cache (0.10× input rate).
• cache_creation – Tokens written to the prompt cache (1.25× input rate).

Returns

Cost in USD for this turn.