# Business Central On-Prem AI Assistant Sizing and Value Calculator

> Source: https://netray.co/tools/business-central-on-prem-ai-assistant-sizing-and-value-calculator
> Publisher: Netray (netray.co), AI and integration engineering for ERP and manufacturing systems.
> Section: Free Tools

## Short answer

Size GPUs, index and monthly value for a private AI assistant on Business Central on-prem from user and question volumes.

## Category

on-prem-ai

## Erp

Microsoft Dynamics 365 Business Central (on-premises)

## Audience

- CIO
- IT manager
- ERP architect
- Finance

## Tool

**Kind:** calculator

**Currency:** $

**Inputs:**

#### Named Business Central users

**Key:** users

**Type:** number

**Default:** 120

**Min:** 1

**Step:** 10

**Unit:** users

#### Share active on a given day

**Key:** dau_pct

**Type:** slider

**Default:** 30

**Min:** 5

**Max:** 100

**Step:** 5

**Unit:** %

#### Questions per active user per day

**Key:** q_per_user

**Type:** number

**Default:** 12

**Min:** 1

**Step:** 1

**Unit:** questions

#### Prompt plus answer tokens per question

**Key:** tokens_per_q

**Type:** number

**Default:** 4000

**Min:** 200

**Step:** 100

**Unit:** tokens

**Help:** API results and retrieved passages add to the prompt.

#### Documents to index

**Key:** docs

**Type:** number

**Default:** 120000

**Min:** 0

**Step:** 10000

**Unit:** documents

#### Average pages per document

**Key:** pages

**Type:** number

**Default:** 3

**Min:** 1

**Step:** 1

**Unit:** pages

#### Minutes saved per question

**Key:** mins_saved

**Type:** number

**Default:** 4

**Min:** 0

**Step:** 1

**Unit:** min

#### Loaded hourly cost

**Key:** hourly

**Type:** number

**Default:** 44

**Min:** 0

**Step:** 1

**Unit:** per hour

#### GPU monthly cost

**Key:** gpu_month

**Type:** number

**Default:** 2400

**Min:** 0

**Step:** 100

**Unit:** per GPU

#### Sustained tokens per second per GPU

**Key:** tps_gpu

**Type:** number

**Default:** 1800

**Min:** 100

**Step:** 100

**Unit:** tokens/s

#### One-time build cost

**Key:** build_cost

**Type:** number

**Default:** 55000

**Min:** 0

**Step:** 5000

**Unit:** one-time

**Outputs:**

#### Daily active users

**Key:** active

**Expr:** users * dau_pct / 100

**Format:** decimal

#### Questions per day

**Key:** q_day

**Expr:** active * q_per_user

**Format:** integer

#### Tokens per day

**Key:** tokens_day

**Expr:** q_day * tokens_per_q

**Format:** integer

#### Peak concurrent requests

**Key:** peak_conc

**Expr:** q_day * 0.2 / 3600 * 20

**Format:** decimal

#### GPUs needed (before redundancy)

**Key:** gpus

**Expr:** max(1, ceil(q_day * 0.2 * tokens_per_q / 3600 / tps_gpu))

**Format:** integer

**Primary:** true

#### Retrieval index size

**Key:** index_gb

**Expr:** docs * pages * 2 * 4096 * 1.5 / 1000000000

**Format:** decimal

#### Hours saved per month

**Key:** hours_month

**Expr:** q_day * 22 * mins_saved / 60

**Format:** hours

#### Monthly value of time saved

**Key:** value

**Expr:** hours_month * hourly

**Format:** currency

#### Monthly net value after GPU cost

**Key:** net

**Expr:** value - gpus * gpu_month

**Format:** currency

**Primary:** true

#### Simple payback

**Key:** payback

**Expr:** if(gt(net, 0), build_cost / net, 0)

**Format:** months

**Chart:**

#### Monthly net value vs questions per active user

**Type:** line

**X:** q_per_user

**From:** 2

**To:** 50

**Steps:** 25

**Series:**

- net

**Disclaimer:** Planning estimate. Measure throughput on your hardware, plan redundancy and validate time savings with a pilot.

## Content

**Intro:** This calculator estimates the hardware and value of a private AI assistant on Business Central on-premises. Enter users and question volume, and it returns GPUs, tokens, index size, hours saved and payback.

**Sections:**

#### How to use it

**Body:** Start with named users and the share who will use the assistant on a typical day. Questions per user depends on how often people look up orders, items and balances. Because API results are added to the prompt, tokens per question often run higher than for plain chat. Replace the default throughput with a measured value from your model and card.

**Bullets:**

- Use pilot data when you have it
- Raise tokens for document questions
- Compare one-GPU and two-GPU outcomes
- Keep the 22-day month in mind

#### The formulas

**Body:** Questions per day equal active users times questions each. Peak tokens per second equal 20% of daily questions times tokens per question divided by 3600. GPUs equal that figure divided by throughput per GPU, rounded up. Index size equals pages times two chunks times 4096 bytes with 50% overhead. Net value equals hours saved times hourly cost minus GPU cost.

#### Reading the result

**Body:** Look at the chart to see how value moves with usage: it rises with questions until a GPU step adds cost. If net value stays low, consider whether built-in reports and search cover the same questions. Payback shows how quickly the one-time build is recovered at the chosen inputs.

**Cta:** If you want these inputs tested against your Business Central environment, Netray can run a short sizing review with your IT team.

**Faq:**

#### Does API use change the sizing?

**Answer:** It affects tokens, because returned records are added to the prompt, and it adds load on the server instance. Keep result sets small and filtered.

#### Why does the value chart have steps?

**Answer:** GPU count is a whole number, so cost jumps when load crosses a capacity limit. The net value line shows those jumps as dips.

#### What default throughput is realistic?

**Answer:** It varies widely by model, quantisation and card. The default is a placeholder for planning; measure your own.

#### Is the index size the full storage need?

**Answer:** No. It covers vectors and metadata only. Keep the original documents and logs in your capacity plan too.

## Search terms

Business Central AI sizing, Business Central on-prem LLM GPU, AI assistant ROI Business Central, private AI payback calculator, LLM tokens per day calculator, retrieval index size estimate

---
Talk to an engineer: https://netray.co/contact?intent=erp_ai | Machine catalog: https://netray.co/agent/catalog.json