Nuvrix AI Cost Review
Nuvrix Pty Ltd · Finance & Accounting
Certification per AWS Marketplace.
Evidence tier Source Confirmed · 4 captures on record
What the publisher says
As described on AWS Marketplace.
**Overview**
The Nuvrix AI Cost Review is a fixed-scope engagement that analyses your AWS AI and machine learning spend and hands you a concrete, prioritised plan to reduce it. It covers token-level spend, model selection, inference configuration, and cost allocation - the levers most teams have never pulled because they didn't know they existed.
Show the rest of the publisher’s description (14 more lines)
AI spend on AWS grows fast and is poorly understood by most finance and engineering teams. Token costs are rarely allocated to the features or workflows driving them. Model choices made during a proof of concept stay in production long after cheaper, faster alternatives became available. Prompt caching and batch inference go unconfigured because nobody flagged the savings opportunity. This engagement surfaces all of it.
**What we analyse**
We pull your cost and usage data at the token level and analyse spend by model, by workload, and by feature where tagging allows. We identify model right-sizing opportunities - cases where a smaller or newer model delivers equivalent output at lower cost. We find prompt caching candidates and quantify the saving. We identify workloads suitable for batch inference. We review your token allocation and tagging to determine what percentage of AI spend is currently unattributable.
**What you receive**
- Token-level spend breakdown by model and workload
- Model right-sizing recommendations with estimated savings
- Prompt caching and batch inference opportunities with effort and saving estimates
- Tag hygiene recommendations to make AI spend fully attributable
- Prioritised savings plan with total identified savings
- Findings session with your engineering and finance leads
**Who it's for**
Organisations running AI workloads on AWS - particularly those using Amazon Bedrock or deploying multiple models in production who want to understand and reduce their AI infrastructure spend. Suited to any team whose AI bill has grown faster than expected and whose finance team is starting to ask questions.
**How it works**
The engagement runs over one week. We access your cost and usage data using read-only permissions, run our query pack against your billing data, and spend the majority of the engagement on analysis and prioritisation. We close with a findings session walking through the savings plan and a clear view of which actions your team can implement immediately.
Highlights
Highlighted by the publisher on AWS Marketplace.
Token-level spend analysis - cost broken down by model, workload, and feature so you know exactly where your AI budget is going and which workloads are driving it.
Savings guarantee - identified annualised savings of at least twice the engagement fee or the fee is halved. With prompt caching, batch inference, and model right-sizing on the table, the bar is low.
Prompt caching and batch inference included - two of the highest-impact cost levers on Bedrock, quantified and prioritised so your team knows exactly what to implement first.
Agent build and provenance
See the full provenance
The layer-by-layer build, the evidence behind each claim, the risk basis and the cross-marketplace links are open to any account. Some rows are disclosed, some the source leaves Unknown; a free account shows you which.
Compliance
- FedRAMPConfirmedNot listed90%, registry-checkedNo FedRAMP Marketplace entry matched this vendor's domain, checked 2026-08-27registry recordas observed 2026-08-27
Confirmed means matched to a public authoritative registry. Claimed means the vendor or its listing states it, not yet cross-checked. A framework not shown was not found in any source we hold, which is not evidence against it. Not listed means a scoped registry check found no match for this vendor's domain: a No is a scoped registry check, not a compliance judgment. Confidence bands: 95% domain-verified, 90% registry-checked, 80% self-attested, 70% weak signal. Self-attested items marked “vendor's site” are gathered from the vendor's own website and are not verified by us.
Sources
Publisher resources
1 linkLinked repositories
Unknown means this listing does not publish a repository. It is not a statement that the code is closed, and a linked repository is not a claim that the publisher wrote it: the registry computes that relationship privately and does not publish it.
Evidence risk is the share of the build you cannot see before you deploy, not a security rating. Sign in to see the layer-by-layer basis for this band.