Prompt Architect
Engineer your prompts with precision
Model roster verified: 2026-07-09
Budget 0.0%
+10 Overhead · Estimatedtokens 0 / 1 048 576
Est. Cost (in) <$0.0001
Est. Cost (out~) $0.0060 Output: ~500 tokens (Medium ≈500)
This model may generate hidden reasoning tokens billed as output but never shown in the response — actual output cost can be several times the estimate above. This tool can't count them; they're decided at generation time.
Expected output length
Prompt Caching
First call $0.0060
Cached call $0.0060
Savings on cached calls Saves 0% vs. first call
This provider caches repeated prefixes automatically server-side — no export markers are needed for the savings above to apply.