Analect Forecast
Analect Forecast prices each AI workload across around 200 models from 27 providers, under your rules on data residency and approved vendors, and gives the reason for every pick. It sets seat licences against the same vendor's pay-per-use pricing and turns the result into a budget.
Part of the AI Spend Forecast Last reviewed by George McKenna, Chairman & CTO, ETT.
Which models fit this work, and what will running it cost?
What goes in
- Usage or billing exports from your AI providers, read in your browser and never uploaded.
- A plain description of each workload: what it does, how often it runs, how large each request is.
- Your seat licences: vendor, seats and price.
- Your rules: data residency, approved vendors, anything that must stay in-house.
What comes out
- A model for each workload, with the reason it won and the models it beat.
- Seats against tokens for each licence, with the break-even point.
- A monthly and annual run-rate, and what it becomes at half and double the volume.
- The date of the price catalogue on every figure.
- A report as PDF, and the estimate as a file.
How is it measured?
- Every figure is labelled as measured from your bills or estimated from an assumption that is stated.
- List prices come from a dated catalogue. Your negotiated rates come from your own bills.
- Seats and tokens are compared with the vendor held constant, so the comparison is a fair one.
What does Analect Forecast not do?
- It does not sit in the path of your AI requests or direct live traffic. It plans; it does not run.
- It does not switch models for you. Where a cheaper model fits the job, it says what to test before you move anything.
- It does not price self-hosted models. It covers models sold per token and per seat.
Where does Analect Forecast fit in an AI pipeline?
Beside it, not inside it. Analect does not sit in the path of your AI requests, and you do not change your models or tools to use it. Analect Forecast works at the point marked below.
- 01
Before you build
What the work should cost, and on which model.
Analect Forecast works here.
- 02
Prepare the documents
What is in them, what they cost to hold, who may see them.
- 03
Choose what the model is given
Whether sending less still gets the same answers.
- 04
Run
Your own models and tools. Nothing of Analect's sits here.
- 05
Check the output
Whether each claim is supported by its source.
- 06
Prove it over time
Evidence an auditor can follow, for one system over a period.
How do we start using Analect Forecast?
Most customers meet Analect Forecast through the AI Spend Forecast, a three-week, fixed-price service ETT delivers on it. Accounts are set up by ETT, and each account sees only the workflows its work needs.
Also under what it costs: Analect Compare. All workflows
About Analect Forecast
Is Analect Forecast the same as the AI Spend Forecast?
Analect Forecast is the workflow. The AI Spend Forecast is the three-week, fixed-price service ETT delivers on it, which ends in a report and a budget your finance team can sign off.
Does Analect Forecast need our prompts or documents?
No. It needs billing or usage exports, your seat licences and a short description of each workload. If you want one prompt priced you can paste it in; it is counted, not stored.
How current are the prices?
They come from a dated catalogue of around 200 models from 27 providers, and the date is on every figure, because a price without a date cannot be checked.
See it on worked material, then on yours
Thirty minutes with the team that runs Analect: the AI Spend Forecast and the Accuracy Evidence File on worked examples, and what either would look like on your bills or your outputs.