01
pre-training
Trained on 4,000+ real-life cycling tests and >100M material configurations, the model has learned to process data recorded as part of a routine battery charging process.
SOH Estimation
From unscalable full-cycle testing to dynamic rapid-charge analysis
Accurate SOH estimation is critical for asset valuation and operational safety — yet traditional testing tools struggle to map non-linear battery degradation, fail on dynamic internal resistance, and can't scale beyond full-cycle lab constraints.
- 4,000+Real-life cycling tests
- < 2%SOH estimation error
- 5xFaster asset insights
- >100MMaterial configurations analyzed
Your asset protection teams are relying on deceptive data readouts
Relying solely on standard Battery Management System (BMS) readouts fails to capture complex, non-linear degradation paths
This is not an engineering failure. It is a process failure.
vetta
One model. All degradation knowledge.
vetta combines standard BMS operational logs, historical fleet averages, and dynamic rapid charge measurements into one flexible battery health framework. Asset operators can instantly unlock independent, lab-grade estimation accuracy and monitor asset state-of-health without interrupting vehicle or grid services.
Three steps from raw charging data
to battery intelligence
02
fine-tuning
Augment model capabilities with historical operational data and additional use case-specific inputs. Develop a custom customer interface to monitor and estimate the SOH of independent assets.
03
inference
Feed real-time operational data to obtain instantaneous SOH predictions as well as expected aging curves. Benchmark independent assets and obtain rich comparative performance information.
What your team
can do with it
01
electric vehicle ecosystems
The model establishes a firm baseline by contextualizing standard BMS readouts and vehicle mileage against historical fleet averages. It then refines this estimate using time-series data from a rapid, 10-minute dynamic voltage check.
02
grid-level energy storage systems
Leverage continuous operational logs, standard BMS data, and short charging sequences to accurately monitor the health of specific modules without interrupting active grid services, ensuring grid stability and preventing thermal events.
03
"zero-to-one" model deployment
Allows OEMs and vehicle diagnostic companies to roll out functional, highly predictive SOH estimation tools for completely new vehicle platforms immediately, without waiting years to
04
few-shot field calibration
Drops the operational SOH estimation error from 5% down to a highly precise 2.5% via full-cycle data modalities, ensuring complex warranty decisions and asset valuations are based on mathematically rigorous truths.
how we deploy
Not a pilot. A permanent
operating relationship.
01
proof of concept
Tested on a sample of your enterprise data — AI-assisted SOH estimations on your historical data arrays with automated insight generation for fleet benchmarking.
Payback in weeks, not months
One unified battery intelligence model trained on your historical asset logs, deployed
securely inside your own infrastructure.
real-time SOH estimation
continuous preventive maintenance
precise field diagnostics