Zur-lix · Effective Cost Reduction

Sign in to view your savings

Enter your API key to unlock cache-aware effective cost reduction for your account.

Effective Cost Reduction

—

Across compression and provider cache, this window.

Estimated cost without us $—

Actual billed cost $—

Estimated savings $—

Total Customer Savings

$—

Compression + Cache + Routing, this window.

Compression Savings

$—

Raw token reduction before the request reaches the provider.

Cache Savings

$—

Provider billing discount on cached prefix tokens, net of any cache-creation overhead. OpenAI may report $0 until automatic cached-token capture ships.

Routing Savings (estimated)

$—

Estimated model-selection savings from routing decisions in the same window.

Estimated routing savings

$—

Projected model-selection savings from the routing ledger.

Actual routing savings

Not yet available

Realized post-request savings are not yet measured for this metric.

Status

request-level join key required

A per-request join key between routing decisions and usage events is required before actual routing savings can be reported.

Compression Savings

—

Tokens removed before the request hit the model.

Cache Hit Ratio

—

Share of input served from the provider's prompt cache.

Input Cost Reduction

—

Input-only economic delta. Diagnostic, not the headline.

Requests

—

— avg latency

Cost before vs. after

Per-bucket dollar cost across the selected window.

Cache multiplier effect. How much your provider's prompt cache amplifies the per-request compression number. A multiplier of 4× means every 1% of compression compounds into ~4% effective savings on this workload.

—