token-budget-pool

Shared token + dollar budget across N concurrent LLM tasks. Thread-safe, rejects records that would push past a cap (totals unchanged on rejection), zero deps.

Usage

use token_budget_pool::{BudgetPool, Caps};
use std::sync::Arc;

let pool = Arc::new(BudgetPool::with_caps(Caps {
    max_input_tokens: Some(10_000),
    max_output_tokens: Some(5_000),
    max_total_tokens: None,
    max_cost_usd: Some(1.0),
}));

// Hand a clone to every task that spends from the budget.
let p = pool.clone();
std::thread::spawn(move || {
    if let Err(e) = p.record(1_000, 500, 0.05) {
        eprintln!("budget exceeded: {e}");
    }
});

License

MIT or Apache-2.0.