Fig. 2 credit model / URL state
FIG. 2
Run the calculator
The Answer Engine side uses live billing-plan pricing when the API is available and the checked-in pricing fallback during static builds. The context-stuffing side is only input-token spend for the average context you resend.
RESULT Pricing source: fallback
MONTHLY DELTA Context higher by $371
Context estimate $1,800 vs memory estimate $1,429.
Fig. 3 not a benchmark / just arithmetic
FIG. 3
Assumptions
The calculator is intentionally conservative and explicit. It does not price output tokens, cache hits, vector infrastructure, engineering time, or accuracy. Add those outside this page when your decision memo needs them.
Context-stuffing estimate
Monthly context cost = queries x average context tokens x input price per 1M tokens / 1,000,000.
Answer Engine estimate
Memory cost = queries x 2 credits x the selected plan's effective cost per credit.
Credit actions
The memory estimate includes one Agent search query (1 credit) and one Agent data retrieval (1 credit) per query from CREDIT_COSTS.
Share state
The result URL stores only queries, tokens, model price, and plan. It does not send source text, account data, or prompts anywhere.
Fig. 4 cost guide / pricing
FIG. 4
Where should you go next?
Use the calculator beside the cost guide and pricing page so the dollar model, technical model, and package model stay aligned.
Fig. 5 FAQPage schema
FIG. 5
FAQ
What does the memory cost calculator compare?
It compares the estimated input-token cost of resending context into an LLM with the Answer Engine credit cost of one agent search query plus one agent data retrieval per query.
Where does the Answer Engine credit price come from?
The page calls the same billing-plan loader as the pricing page, which fetches GET /api/v1/billing/plans when available and falls back to the checked-in pricing model when the API is unavailable during static builds.
Is the result a benchmark or savings promise?
No. The result is arithmetic from the inputs you choose. It does not claim that Answer Engine is always cheaper, more accurate, or autonomously improving.
Why does the calculator ask for input-token price?
Provider pricing changes by model and vendor, so the calculator lets you enter the input-token price shown on your provider pricing page instead of baking in one model.