What is the AI Decision Platform?

A configuration-driven layer for AI planning tools: model registry, pricing registry, and shared decision engines.

Are prices guaranteed?

No. Prices are planning estimates with confidence badges and official source links. Verify before production budgeting.

Will prices auto-update?

Collector interfaces are prepared for future sync (OpenAI, Anthropic, Google, OpenRouter, xAI, DeepSeek). Scraping is intentionally not implemented.

AI Decision Platform

面向建構者的 AI 決策引擎

Plan product cost, choose stacks, and size context windows using a shared model catalog and pricing registry—not one-off token price widgets.

Product 2026.07.5 · Catalog 2026.07.2 · Pricing pricing-2026.07.2 · Trust 2026.07.3 · Decision 2026.07.4 · Core 1.0.0

AI 產品成本規劃器

Estimate AI + infra opex, cost per user, suggested price, break-even, and health score.

開啟規劃器

AI 技術棧規劃器

Recommend models by product type, budget, latency, and quality priorities.

開啟規劃器

上下文規劃器

Estimate tokens, context usage, chunk size, and whether a corpus fits.

開啟規劃器

Model providers in the registry

Models and prices live in configuration registries with confidence levels and official URL sources—ready for future collectors without rewriting planners.

OpenAI
Anthropic
Google
xAI
DeepSeek
OpenRouter

Frequently asked questions

What is the AI Decision Platform?: A configuration-driven layer for AI planning tools: model registry, pricing registry, and shared decision engines.
Are prices guaranteed?: No. Prices are planning estimates with confidence badges and official source links. Verify before production budgeting.
Will prices auto-update?: Collector interfaces are prepared for future sync (OpenAI, Anthropic, Google, OpenRouter, xAI, DeepSeek). Scraping is intentionally not implemented.

From calculators to AI decision engines

Builders need multi-variable decisions: workload, margin, latency, quality, and context fit. CalcRecipe’s AI platform encodes those decisions on shared registries.

Future collectors, historical pricing, and workshop notes can plug into the same catalog without rewriting planner UI.

Version history

2026.07.2
2026-07-20
Added OpenAI GPT-5.x models observed in an authenticated API Platform limits table. Availability is account-scoped; token prices remain unverified until an official pricing source is confirmed.
Builders need current API model IDs without inventing prices from rate limits.
2026.07.5
2026-07-20
CalcRecipe Core v1.0 — The platform foundation has been stabilized. Builder tools now share transparent pricing sources, explainable recommendations, visible assumptions, confidence and trade-off analysis, and reproducible version snapshots. This release prepares CalcRecipe for new specialized decision tools.
Freeze shared decision, pricing-trust, and version contracts so future domains extend through adapters.
2026.07.4
2026-07-20
Introduced the domain-agnostic Decision Engine Core; migrated AI Stack Planner to explainable weighted scoring.
Builder recommendations must be reproducible, constraint-aware, and explainable across domains.
1.1.0
2026-07-20
AI Stack Planner v1.1 — Decision Engine adapter with priority presets, confidence, sensitivity, and policy snapshots.
Replace opaque heuristic scores with a versioned decision policy.
2026.07.3
2026-07-20
Added the AI Pricing Trust Layer with user overrides, freshness states, and transparent source labels.
Catalog pricing can be incomplete or stale; builders must inject real rates without mislabeling them as verified.