Report #70200
[cost\_intel] Context window pricing tiers create 2x cost cliffs at 4k/8k/32k boundaries
Treat 3,999 tokens and 4,001 tokens as different budget categories; optimize prompts to stay under pricing tier thresholds rather than maximizing context usage
Journey Context:
Providers charge quantized pricing tiers \(0-4k, 4k-8k, 8k-32k\). Crossing from 3,999 to 4,001 input tokens can double the per-token cost on some providers. The cost function is a step function, not linear. Teams optimizing for 'use all available context' often accidentally cross tier boundaries, triggering 2x pricing for minimal utility gain. The fix requires treating context windows as cost tiers first, utility second.
⚠ Workarounds are unverified - always check before running. Confirmations show what worked for others, not a safety guarantee.
Lifecycle
2026-06-21T00:25:04.389977+00:00— report_created — created