Methodology
The forecast multiplies users, calls, days, and token sizes, then applies input and output prices independently.
Token and API Costs
Forecast organization-wide model spend from users, calls, token sizes, workdays, and model prices.
Interactive Tool
| Result | Value | How to read it |
|---|---|---|
| Organization calls / month | 44K | Users times daily calls times active days. |
| Input tokens / month | 66M | Monthly prompt and context demand. |
| Output tokens / month | 17.6M | Monthly generated demand. |
| Monthly model spend | $1,540.00 | Lowest known token prices for the selected model. |
| Annualized run rate | $18,480.00 | Twelve times the selected monthly scenario. |
Separate heavy agents, batch jobs, images, audio, embeddings, storage, and retries when they materially change the average.
Catalog model: Anthropic Claude Fable 5 →
Inputs
| Input | How it is used |
|---|---|
| Users | People or active product users generating calls. |
| Calls per user | Average successful calls per active day. |
| Tokens per call | Average billed input and output tokens. |
| Days and model | Active days and source-linked model rates. |
The forecast multiplies users, calls, days, and token sizes, then applies input and output prices independently.
Create separate scenarios for heavy and light users when the distribution is skewed. A single average can understate high-cost agents or batch jobs.
Boundaries
Catalog values retain their source and freshness on the linked model, provider, benchmark, or comparison page. Editable scenario assumptions are not Model Markets measurements.
Continue the Analysis
| Tool | Next question |
|---|---|
| API Usage Calculator | Translate request-level product assumptions into monthly token demand and spend. |
| Coding Agent Cost | Convert an agent's observed hourly token burn into a monthly cost scenario. |
| API vs Subscription | Find the usage level where seat pricing and metered API pricing cross for a team. |
Questions
Build a bottom-up monthly AI budget from product or team behavior. It returns organization requests, token demand, per-user cost, monthly total, and annualized run rate.
Where the calculation needs model facts, it uses the current Model Markets catalog snapshot updated 2026-09-02. User-entered assumptions remain clearly editable, and unsupported values stay unknown rather than being inferred.
Non-token products, images, audio, storage, and vector services are excluded. Retries and failed calls must be included in call assumptions. Annualized run rate assumes the selected month is representative. Open the linked canonical records and primary sources before making a production or purchasing decision.