Module cost_estimator 

Pre-flight cost estimation for LLM requests.

Provides token-count heuristics, per-model pricing, and budget-checking before a request is dispatched so that callers can decide whether to proceed, choose a cheaper model, or reject the request entirely.

Structs

BudgetConfig

Spending limits used to check whether a request is within budget.

CostEstimate

Full cost estimate including budget compliance.

CostEstimator

Pre-flight cost estimation engine.

TokenEstimate

Input/output token count estimate for a request.

Enums

TaskType

Characterises the kind of work being requested, which influences the expected output token count.