# SAP Business One AI assistant sizing and value calculator

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

## Short answer

Estimate peak concurrent requests, GPUs, tokens per day, index size, hours saved and payback for a private AI assistant on SAP Business One.

## Category

sap

## Erp

SAP Business One

## Audience

- CIO
- IT architect
- Finance controller

## Tool

**Kind:** calculator

**Currency:** $

**Inputs:**

#### Named users

**Key:** users

**Type:** number

**Default:** 40

**Min:** 1

**Step:** 10

**Unit:** users

**Help:** Users with access to the assistant.

#### Daily active share

**Key:** active_pct

**Type:** slider

**Default:** 50

**Min:** 5

**Max:** 100

**Step:** 5

**Unit:** %

**Help:** Share of named users who ask at least one question on a typical day.

#### Questions per active user per day

**Key:** qpd

**Type:** number

**Default:** 10

**Min:** 1

**Step:** 1

**Unit:** questions

#### Average prompt plus answer size

**Key:** tokens

**Type:** number

**Default:** 2000

**Min:** 200

**Step:** 100

**Unit:** tokens

**Help:** Includes retrieved context. Retrieval-heavy answers often run 2,000 to 6,000 tokens.

#### Average seconds per answer

**Key:** resp_sec

**Type:** number

**Default:** 20

**Min:** 2

**Step:** 1

**Unit:** s

**Help:** End to end, including retrieval and generation.

#### Documents to index (attachments, SOPs, quotes)

**Key:** docs

**Type:** number

**Default:** 15000

**Min:** 0

**Step:** 1000

**Unit:** docs

#### Average pages per document

**Key:** pages

**Type:** number

**Default:** 4

**Min:** 1

**Step:** 1

**Unit:** pages

#### Minutes saved per question

**Key:** mins

**Type:** number

**Default:** 5

**Min:** 0

**Step:** 1

**Unit:** min

**Help:** Be conservative: time to find the answer today minus time with the assistant.

#### Loaded hourly cost

**Key:** hourly

**Type:** number

**Default:** 30

**Min:** 1

**Step:** 1

**Unit:** $/h

#### Aggregate tokens per second per GPU

**Key:** tps

**Type:** number

**Default:** 600

**Min:** 50

**Step:** 50

**Unit:** tok/s

**Help:** Batched serving throughput of your model on your GPU. Measure with your own model; this default is a planning figure.

#### Monthly cost per GPU

**Key:** gpu_cost

**Type:** number

**Default:** 900

**Min:** 0

**Step:** 100

**Unit:** $/month

**Help:** Amortised hardware, power and support, or your private cloud rate.

#### One-time setup cost

**Key:** setup

**Type:** number

**Default:** 35000

**Min:** 0

**Step:** 5000

**Unit:** $

**Help:** Integration, security review, indexing and pilot work.

**Outputs:**

#### Daily active users

**Key:** active_users

**Expr:** users * active_pct / 100

**Format:** integer

#### Questions per day

**Key:** queries_day

**Expr:** active_users * qpd

**Format:** integer

#### Tokens per day

**Key:** tokens_day

**Expr:** queries_day * tokens

**Format:** integer

**Description:** Questions per day times average tokens per question.

#### Peak concurrent requests

**Key:** peak_conc

**Expr:** ceil(queries_day * 0.2 / 3600 * resp_sec)

**Format:** integer

**Description:** Assumes 20% of daily questions fall in the busiest hour.

#### GPUs for peak load

**Key:** gpus

**Expr:** max(1, ceil(queries_day * 0.2 * tokens / 3600 / tps))

**Format:** integer

**Primary:** true

**Description:** Peak-hour tokens per second divided by aggregate tokens per second per GPU.

#### GPUs with one spare

**Key:** gpus_ha

**Expr:** gpus + 1

**Format:** integer

**Primary:** true

**Description:** Adds one GPU for failover and maintenance.

#### Retrieval index size

**Key:** index_gb

**Expr:** docs * pages * 3 * 4096 * 1.5 / 1000000000 + docs * pages * 2500 / 1000000000

**Format:** decimal

**Description:** About 3 chunks per page, 1024-dimension float32 vectors with 1.5x index overhead, plus roughly 2.5 KB of text per page. In GB.

#### Hours saved per month

**Key:** hours_month

**Expr:** queries_day * mins / 60 * 21

**Format:** hours

**Description:** 21 working days per month.

#### Monthly value of time saved

**Key:** monthly_value

**Expr:** hours_month * hourly

**Format:** currency

#### Monthly GPU cost

**Key:** monthly_cost

**Expr:** gpus_ha * gpu_cost

**Format:** currency

#### Net monthly value

**Key:** net

**Expr:** monthly_value - monthly_cost

**Format:** currency

#### Simple payback

**Key:** payback

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

**Format:** months

**Description:** Months to recover the setup cost. Zero means the monthly value does not cover GPU cost.

**Chart:**

#### Monthly value vs GPU cost by named users

**Type:** line

**X:** users

**From:** 10

**To:** 200

**Steps:** 30

**Series:**

- monthly_value
- monthly_cost

**Disclaimer:** Planning estimate only. Measure real throughput with your model and hardware, and validate time saved with a pilot before committing budget.

## Content

**Intro:** This calculator estimates how many GPUs, how much index storage and what monthly value a private AI assistant on SAP Business One is likely to need. Enter your user counts, question volumes and costs; it returns peak concurrency, GPU count, hours saved and payback, with a chart sweeping named users.

**Sections:**

#### How to use it

**Body:** Start with named users and the share who use the assistant on a typical day, then estimate questions per active user. For documents to index (attachments, sops, quotes), count what you really intend to index, not everything on the file server. Set tokens per answer from a prototype if you have one; retrieval context makes answers longer than chat. The tool is aimed at SAP Business One cases like Business One companies with a few dozen users and a single GPU. Change one input at a time and watch the chart.

**Bullets:**

- Use measured tokens per second for your own model
- Keep minutes saved conservative
- Remember the spare GPU for failover
- Revisit after the pilot with real usage data

#### The formulas

**Body:** Daily active users equal named users times the active share. Questions per day equal active users times questions per user. Tokens per day equal questions times average tokens. Peak concurrent requests equal questions per day times 0.2 (the busiest-hour share) divided by 3,600, times seconds per answer, rounded up. GPUs equal peak-hour tokens per second divided by aggregate tokens per second per GPU, rounded up, with a minimum of one; the calculator then adds a spare. Hours saved per month equal questions per day times minutes saved divided by 60, times 21 working days.

#### Reading the result

**Body:** If net monthly value is negative, the user base or question volume is too low to justify dedicated GPUs, and a smaller model, shared hardware or a narrower use case is the better answer. If payback is under a year with conservative minutes saved, the case is reasonable. Index size is usually small compared with model weights; the model and its GPU memory dominate hardware planning. Treat value as an upper bound until a pilot shows people actually save the time.

#### Common mistakes

**Body:** Using the licensed user count instead of real daily users inflates GPUs. Counting every question as saving the same time overstates value, because many questions replace a thirty-second lookup. Using vendor peak throughput numbers measured on short prompts understates GPU need for long retrieval prompts. And ignoring the one-time work of permissions, indexing and security review makes payback look better than it is.

**Cta:** If you want these numbers checked against a real pilot on SAP Business One, Netray can run a scoped proof of value on your own hardware.

**Faq:**

#### How do I find tokens per second per GPU?

**Answer:** Benchmark your chosen model on your GPU with a serving stack such as vLLM, using realistic prompt lengths and many concurrent requests. Use aggregate throughput, not single-stream speed. The default in this tool is a planning figure only and will vary widely by model size and quantisation.

#### Why add a spare GPU?

**Answer:** A single GPU is a single point of failure, and you also need capacity during maintenance and model updates. The spare is a simple allowance. Some sites accept degraded service on failure and skip it; set the cost input accordingly if you do.

#### Is the 20% busiest-hour share right?

**Answer:** It is a reasonable planning default for a single-site working day. Multi-shift or multi-timezone sites spread load more evenly and need fewer GPUs; sites with a morning rush may need more. Adjust by scaling questions per day or adding headroom.

#### Does this include the cost of people?

**Answer:** Only through the one-time setup input, which you should fill with integration, security review and pilot effort. Ongoing support and model upgrades are not modelled, so add a margin to the monthly GPU cost if you expect meaningful run effort.

## Search terms

SAP Business One AI sizing, on-prem LLM GPU sizing calculator, AI assistant ROI SAP Business One, how many GPUs for internal LLM, LLM tokens per day calculator, on-prem AI payback calculator

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