What you get, and how good it is — in one place. The Plans tab is the tiers, SLAs, and quotas the gateway enforces; the Benchmark tab is our measured routing quality and the verified-fix receipts, never vendor benchmarks and never placeholder numbers.
Super Intelligence Dog is the engine that does the work; BellerBuild is the proof layer that scans the result. Plans are billed by BellerCreatives — the account of record — and every SID key is issued under that one account. Monitoring an app you have already shipped? See BellerBuild plans.
Loading…
Three numbers per task class, measured on our own versioned task set — never vendor benchmarks, never live customer traffic. Cost per passing result is the arm's total metered cost on that class divided by its passing runs — failed runs are priced in, not hidden. The arms: one pinned open-weight model vs SID auto routing, at the same task-set version and seed.
Loading…
Loading…