{"product_id":"tokens-ai-cost-planner","title":"Tokens — AI Cost, Token \u0026 Context Budget","description":"\u003cp\u003e\u003cstrong\u003eTokens — AI Cost, Token \u0026amp; Context Budget\u003c\/strong\u003e — a premium, 100% offline digital tool by Aceplore. Runs entirely in your browser: no accounts, no subscriptions, your data never leaves your device.\u003c\/p\u003e\n\u003cp\u003eA private, offline cost bench for anything built on a model — what one run costs term by term, what a thousand of them cost, what the cache is actually saving and what still fits in the window once output is reserved — 31 tools in one, and every rate is a figure you entered.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003e31 built-in tools in one app.\u003c\/strong\u003e A rate card holding three prices per model — input, cached input and output — plus the context window and the max output, every cell editable and none of it prefilled, with a model detail view, a compare screen costing one token profile against every model on the card, and a printable card that records where each figure came from; workloads as a token profile, a model and a volume, a cost-per-run view printing every term with its inputs, its units and its result, and a batch view pricing any number of identical runs; monthly and annual projections, budgets with a cap per group of workloads, a burn chart of logged spend against the pace of the month, and scenarios that freeze the rate card and the workloads so two plans can be compared side by side; a run log recording what actually ran and what it was charged, re-costed at today's rates, and an estimate-versus-actual view reporting drift per call rather than per log entry; cache savings at the cached rate you entered, a breakeven view for the write premium you enter, and an output-share view separating the share of tokens from the share of the bill; a text estimator that divides characters by a ratio and labels the result an estimate, with a calibration screen for measuring that ratio against text whose real token count you already have; context budgeting with the output reserve taken off first, headroom per workload and a truncation order you set; plus a method page listing every formula and every place the app refuses to guess, a printable report, CSV export and settings.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eCost per run.\u003c\/strong\u003e Uncached input, cached input and output are each divided by 1,000,000, multiplied by their own rate and then added. At 3.00 in and 15.00 out per 1M, 6,000 input and 2,000 output tokens cost 0.048 — 0.018 input, 0.030 output — which is why shortening answers moves the bill faster than shortening prompts.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eCache savings.\u003c\/strong\u003e Cache 5,000 of that 6,000-token input at 0.30 per 1M and the run falls from 0.048 to 0.0345, a saving of 0.0135. The cached tokens are billed at their own rate; nothing here assumes a standard fraction of the input price.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eContext budget \u0026amp; Window budget.\u003c\/strong\u003e Reserve 8,000 tokens for output against an 80,000-token window and 72,000 of input room is left, which a 6,000-token prompt leaves 66,000 clear of. Enter no window and the app reports no utilisation figure at all rather than inventing a denominator.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eWhat's included:\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul\u003e\n\u003cli\u003eTokens.html\u003c\/li\u003e\n\u003cli\u003eQuick-Start-Guide.html\u003c\/li\u003e\n\u003cli\u003eHow-To-Use-Guide.html\u003c\/li\u003e\n\u003cli\u003eREADME.txt\u003c\/li\u003e\n\u003cli\u003eLICENSE.txt\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003eIncludes a personalized welcome wizard, an interactive guided tour, a ⌘K command palette, light\/dark themes with 8 accents, and one-click JSON backup \u0026amp; restore.\u003c\/p\u003e\n\u003cp\u003e\u003cem\u003eTokens is a planning, costing and record-keeping tool for people building with AI. Every figure it reports is computed by a stated formula from the rates, sizes, counts and thresholds YOU enter. It ships no model prices, no context windows and no tokeniser: those differ by provider and model and change without notice, so you enter them and the app shows which ones it used. Any token figure derived from the length of a text is an ESTIMATE, not a count — real tokenisers are model-specific and cannot run offline — so treat it as a planning figure and confirm actual usage against your provider's own reporting and billing before you commit to a budget. Nothing here is a guarantee of a cost, a saving, a quality level or a result, and none of it is business, financial, legal or engineering advice. You are solely responsible for your own decisions.\u003c\/em\u003e\u003c\/p\u003e","brand":"Aceplore","offers":[{"title":"Personal","offer_id":51816833581196,"sku":"TOKENSPER","price":39.0,"currency_code":"CAD","in_stock":true},{"title":"Commercial","offer_id":51816833613964,"sku":"TOKENSCOM","price":79.0,"currency_code":"CAD","in_stock":true},{"title":"Extended","offer_id":51816833646732,"sku":"TOKENSEXT","price":149.0,"currency_code":"CAD","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0779\/3763\/9564\/files\/tokens-00-cover.png?v=1785800747","url":"https:\/\/aceplore.com\/products\/tokens-ai-cost-planner","provider":"Aceplore","version":"1.0","type":"link"}