Softwr

AI · head to head

Avathon vs Novity

Avathon logo

Avathon

AI

Industrial AI for asset performance and renewables, formerly SparkCognition

From
On request
Rated
-
Full reviewWebsite
Novity logo

Novity

Energy

Hybrid physics and machine learning prognostics that estimate remaining useful life for process equipment

From
On request
Rated
-
Full reviewWebsite

The short version

  • Each has a real cost: Avathon the company rebranded from SparkCognition to Avathon and narrowed its focus after repeated reported layoff rounds, so reference checks and roadmap commitments made under the old name need reconfirming.; Novity novity is a small venture-backed company with a strategic investor rather than a profitable business, so continuity risk is real and the Tokyo Gas investment signals a likely eventual acquisition that would reset the roadmap.
  • They diverge on capability: Avathon covers Renewable Suite, Novity covers TruPrognostics engine.
  • Prices and features above were last checked on 1 September 2026.

Where they differ

Only the attributes on which Avathon and Novity actually diverge.

Attributes where Avathon and Novity differ
AttributeAvathonNovity
CategoryAIEnergy

Identical on both: starting price (On request), pricing model (quote), free tier (No), platforms (Web, Cloud), user rating (Not yet rated).

What each one covers

Drawn from each product's published feature list. An absence here means we hold no record of it - not that the product lacks it.

Only in Avathon

  • Renewable Suite
  • Loss attribution
  • Predictive maintenance models
  • Computer vision
  • Digital twin models
  • Work order integration
  • Fleet benchmarking
  • Document intelligence

Only in Novity

  • TruPrognostics engine
  • Cold-start modelling
  • Fault mode diagnosis
  • Remaining useful life
  • Existing sensor reuse
  • Recommended actions
  • Historian connectors
  • Asset class libraries

What people use each for

The jobs each tool is most often brought in to do.

Avathon

  • A wind fleet owner wanting underperforming turbines ranked by lost megawatt hours rather than by alarm countnot Novity
  • A solar independent power producer detecting inverter and string underperformance that falls below the SCADA alarm thresholdnot Novity
  • A manufacturer with years of historian data and failure records who wants failure prediction models trained on their own historynot Novity
  • A site using cameras to detect missing personal protective equipment or unauthorised entry into restricted zonesnot Novity

Novity

  • A gas processing plant that needs a defensible time-to-failure number before deferring a turnaroundnot Avathon
  • An LNG terminal with critical compressors and no run-to-failure history to train a conventional modelnot Avathon
  • A wastewater operator whose existing vibration alarms are ignored because they carry no severity or horizonnot Avathon
  • A generator operator supplying data centre load where an unplanned trip carries contractual penaltiesnot Avathon

Where each one falls short

Documented limitations, not opinions. Every one is a constraint you would hit in normal use.

Avathon

  • The company rebranded from SparkCognition to Avathon and narrowed its focus after repeated reported layoff rounds, so reference checks and roadmap commitments made under the old name need reconfirming.
  • Headcount is in the low hundreds against a product portfolio spanning renewables, vision, predictive maintenance and document AI, which is thin coverage per product line for enterprise support expectations.
  • Model accuracy depends on having enough labelled historical failures, and sites with short data history or poor maintenance record-keeping get generic anomaly detection rather than useful failure prediction.
  • Nothing is published about pricing or asset-count thresholds, so there is no way to judge whether a small fleet is even a viable customer without entering a sales cycle.
  • It sits on top of your SCADA and CMMS rather than replacing them, so the integration and asset-hierarchy mapping work is a real project cost that the software subscription does not cover.

Novity

  • Novity is a small venture-backed company with a strategic investor rather than a profitable business, so continuity risk is real and the Tokyo Gas investment signals a likely eventual acquisition that would reset the roadmap.
  • Physics-based models must be configured per equipment class, so each new asset type is an engineering engagement rather than a configuration screen, and rollout speed is limited by Novitys own capacity.
  • Prognostics depend on the quality and sampling rate of your historian data; plants recording ten-minute averages will not get useful remaining-useful-life estimates without new instrumentation.
  • Nothing about pricing is published and there is no self-service entry point, so evaluation always starts with a sales-led pilot on a handful of assets.
  • The deployment footprint is concentrated in oil and gas, LNG and water, so reference customers and pre-built asset models outside those industries are limited.

Pricing, plan by plan

Avathon

On request
  • Avathon$undefined/year
    • Annual subscription sized by asset count or installed megawatts for renewables
    • Separate pricing for vision, predictive maintenance and document products
    • Model development and data onboarding quoted as a professional services project

Novity

On request
  • TruPrognostics$undefined/year
    • Quoted per asset class and monitored equipment count
    • Model configuration and commissioning quoted as a project
    • Typically an annual subscription tied to a pilot then a rollout

Which should you pick?

Choose Avathon if

  • You need renewable suite.
  • You work on Web, Cloud.
  • You also want loss attribution.
Avathon review

Choose Novity if

  • You need truprognostics engine.
  • You work on Web, Cloud.
  • You also want cold-start modelling.
Novity review

Questions people ask

Is Avathon or Novity better?
Neither clearly leads. Avathon starts at On request and Novity at On request, and user ratings are close enough to be indistinguishable. Choose on capability and platform support.
Which is cheaper, Avathon or Novity?
Avathon starts at On request and Novity at On request.
Does Avathon or Novity run on more platforms?
Both run on Web, Cloud, so platform support will not decide this one for you.
What is Avathon best used for?
Avathon is most often used for a wind fleet owner wanting underperforming turbines ranked by lost megawatt hours rather than by alarm count, a solar independent power producer detecting inverter and string underperformance that falls below the scada alarm threshold, a manufacturer with years of historian data and failure records who wants failure prediction models trained on their own history, a site using cameras to detect missing personal protective equipment or unauthorised entry into restricted zones. Of those, a wind fleet owner wanting underperforming turbines ranked by lost megawatt hours rather than by alarm count and a solar independent power producer detecting inverter and string underperformance that falls below the scada alarm threshold are not what Novity is typically brought in for.
What can Avathon do that Novity cannot?
Avathon covers Renewable Suite, Loss attribution, Predictive maintenance models, Computer vision. Novity covers TruPrognostics engine, Cold-start modelling, Fault mode diagnosis, Remaining useful life.

Answered from the vendors’ own pages

Avathon: Is this the same company as SparkCognition?

Yes. SparkCognition rebranded to Avathon and refocused on industrial and government asset AI.

Novity: What does Novity actually output?

A named failure mode and an estimated remaining useful life with a confidence band, not just an anomaly alert.

Avathon: Which product is most mature?

The renewable energy suite for wind and solar underperformance, which has the longest production deployment history.

Novity: Do we need failure history to train it?

No. The physics component is what lets it produce useful prognostics on equipment with little or no run-to-failure data.

Avathon: Does it need my historian data in the cloud?

Generally yes for model training and ongoing scoring, which is a security review item for utility and defence customers.

Novity: Do we need new sensors?

Often not. It reads from your existing historian, but low sampling rates or missing measurements can require additional instrumentation.

Avathon: Can it work without historical failure records?

It will still do anomaly detection, but failure prediction models need labelled historical failures to be worth the deployment cost.

Novity: Who backs the company?

It was spun out of Xerox PARC and took a strategic investment from Acario Innovation, the venture arm of Tokyo Gas, in 2026.

Share

Related pages

More on Avathon

More on Novity

Keep looking

Other head to heads