Sage X3 AI Assistant Sizing and Value Calculator
Size GPUs and estimate value for a private AI assistant on Sage X3. Enter named users, usage and cost assumptions to get concurrency, tokens per day, index size and payback. It uses plain formulas you can check and change.
Your numbers
Share of named users who ask the assistant something on a given day.
Includes retrieved context. Short SQL answers run lower; document questions run higher.
Time the user would otherwise spend finding the answer. Measure it in a pilot.
Amortised hardware plus power and hosting per GPU, or the monthly rate for private cloud GPUs.
Aggregate throughput with batching for your chosen model and quantization. Measure with vLLM on your hardware.
Your results
Planning model. Formulas: questions/day = users x active share x questions each; tokens/day = questions x tokens; GPUs = ceil(peak-hour tokens per second / tokens per second per GPU); value = questions/day x 22 working days x minutes saved / 60 x hourly cost. The 20% peak-hour share and 20 second request time are fixed assumptions. Validate throughput with your own model and prompts.
GPUs needed vs questions per active user
Sweeps Questions per active user per day from 2 to 60, holding your other inputs fixed.
Validate my sizing
Optional: have the numbers checked with a measured pilot on your hardware. The calculator works fully without it.
Optional. The tool and every export work without this. No spam.
How to use it
Enter the number of named Sage X3 users and the share who use the assistant daily. Add questions per active user, average tokens per question, documents and pages for the index, minutes saved, loaded hourly cost and GPU monthly cost. The tokens per second per GPU input is the most important assumption, so replace the default with a number from a test on your hardware.
- Start with a pilot group size and read hours saved.
- Raise tokens per question if you use long documents.
- Use the chart to see how demand scales with users or usage.
The method behind it
Daily tokens are questions times tokens per question. Peak load assumes 20% of daily questions land in the busiest hour, converted to tokens per second and divided by per-GPU throughput. Concurrency uses the same arrival rate times a 20 second request time. Index size assumes about three chunks per page, each around 9 KB for text and vectors. Value is minutes saved converted to hours, times 22 working days and hourly cost.
How to read the result
Typical Sage X3 questions cover order status by site, stock by location, work order progress and open purchase receipts, plus questions that need documents such as certificates or contracts. If the GPU count is one, plan a second for resilience once the assistant is relied on. If payback days exceed 30, either the usage is low, the time saved is small, or the hardware is oversized for the load.
Common mistakes
Counting named users as active users, ignoring retrieved context in the token count, trusting vendor throughput numbers for a different model, and claiming time saved without a pilot measurement. Another is forgetting that model quality, not just speed, decides whether people keep using the tool. Treat value as a hypothesis until a pilot confirms it.
Frequently Asked Questions
Why is tokens per second per GPU an input?
It depends on the model, quantization, context length and batching. Public numbers rarely match your workload, so measure with vLLM on your hardware and enter the sustained figure.
Does the GPU estimate include redundancy?
No. It is the minimum for peak-hour load. For a production assistant, add at least one spare GPU or a failover plan.
How reliable is the value figure?
It is only as good as minutes saved per question. Measure that in a pilot by timing a sample of tasks with and without the assistant.
What does the index size include?
An estimate of text chunks and vectors for the documents you enter. It excludes the ERP database copy, which is sized separately by your DBA.
Want these numbers tested against your real Sage X3 data and prompts? Netray can run a measured pilot on your own GPUs and report throughput and accuracy.
Related Tools
Sage X3 Private AI Layer Blueprint Generator
Generate an architecture plan for adding a private AI assistant to on-prem Sage X3: SQL replica views, Syracuse SData and web services, retrieval, local GPU model, permissions and audit.
On-Prem AIOn-Prem GPU Sizing Calculator for ERP AI
Work out how much GPU memory and how many cards you need to serve an open-weight model to your ERP users on-prem.
On-Prem AIvLLM Throughput Estimator
Estimate aggregate tokens-per-second throughput for a vLLM deployment from model size, GPU class, and batch depth, accounting for continuous batching gains.
On-Prem AIOn-Prem AI ROI Calculator
Turn hours saved per employee into annual net benefit, payback months, and 3-year ROI for an on-prem AI investment.
More Other ERPs tools
See all- Sage X3 Import Template Worksheet
- Sage X3 Database Tables Reference
- Sage X3 Maintenance Checklist and Schedule
- Infor VISUAL Work Order Status and Structure Reference
- Infor VISUAL Manufacturing Scheduling Checklist
- Infor XA (MAPICS) Period End Close Checklist
- Infor VISUAL Database Table Reference
- Infor XA (MAPICS) Application Module Reference
- IBM i Commands for Infor XA Administrators Reference
Keep exploring
Free tools
- Sage X3 Import Template WorksheetAn editable import tracker for Sage X3: which object, which import template, source file, records loaded, errors and owner, with error rates calculated.
- Sage X3 Private AI Layer Blueprint GeneratorGenerate an architecture plan for adding a private AI assistant to on-prem Sage X3: SQL replica views, Syracuse SData and web services, retrieval, local GPU model, permissions and audit.
- Sage X3 Database Tables ReferenceKey Sage X3 tables by module with what they store, typical key columns and naming conventions for SQL and reporting.
ERP fixes and how-tos
- Sage X3 AI Integration On-Prem: A Private ArchitectureHow to add a private AI assistant to on-prem Sage X3: SQL read replica, Syracuse SData and web services, retrieval index, local GPU model and audit trail.
- Sage X3 Production Planning, Scheduling and TrackingHow Sage X3 production planning works: master data, MRP, work orders, scheduling and shop floor tracking, and what to set up first. Practical overview.
- Sage X3 production and warehouse management: how the setup fits togetherHow Sage X3 production and warehouse management fit together: master data, work orders, locations, stock movements and the order to set them up in.
AI on your ERP
- AI for Sage X3, Grounded in Your Own Ledgers and Stock DataAdd AI to Sage X3 finance and stock data without exporting it to a public model. On-prem and private-cloud LLMs for Sage X3 in the UK and France.
- AI for Sage X3 Purchasing and Stock: Orders, Receipts and LotsAI for Sage X3 purchasing and stock: PO review, supplier delivery analysis, stock status and lot questions over X3 data using GraphQL, SOAP and private models.
Integrations
- How to add an AI chatbot to Sage X3Add an AI chatbot to Sage X3 for staff, suppliers or customers: grounded in ERP data, respecting roles, with a fully on-prem option using open models.
- How to add AI agents to Sage X3How to add AI agents to Sage X3: a tool layer over SData and web services, read-first design, human approval on writes, and on-prem models for MRP and POs.