Models
The right modelfor each kind of time series
Each model is built for one kind of time-series data. Pick a model to see what it does, what data it takes, and where it stands today.
General time-series foundation model for energy, transportation, finance and other series, available on the SequenxAI Platform
General time-series foundation model · Available on the SequenxAI Platform
One modelfor the rhythms of many domains
SequenxAI Omni is SequenxAI's general time-series foundation model, available on the SequenxAI Platform. It is built for energy, transportation, finance and other time series, and aims to forecast new, unseen series directly — point forecasts with quantile intervals — as a cold-start baseline that key industries then tune further.
How the model is meant to work Illustrative data
Multi-domain series → general foundation model → quantile forecasts
Structural sketch, not a live forecast.P10–P90
Design goals
What the model is built to do
These are the model's design goals, not verified capability commitments; accuracy conclusions rest on same-scenario comparison results.
- 01
Learning across domains
Aims to learn the trends, cycles and seasonality shared by time series from many different sources, as one starting point across scenarios; industry expertise is added on top through domain tuning.
- 02
Zero-shot forecasting
Forecasts new, unseen series directly, so a new site need not wait months for data; key industries and customer scenarios are then tuned on top.
- 03
Quantile intervals
Outputs point forecasts with quantile intervals such as P10–P90, showing the plausible range of outcomes.
- 04
Verifiable evaluation
Same-scenario comparisons on partners' real data, evaluated in a reproducible way.
Model summary
The model at a glance
- Model type
- General time-series foundation model
- Input modality
- Time series
- Target domains
- Energy · Transportation · Finance and more
- Output
- Point forecast + quantile intervals
- Forecasting mode
- Zero-shot cold start + domain tuning (design goal)
Target Scenarios
Built for high-stakes time-series decisions
Grid Load & Renewable Dispatch
Energy series for load, generation and power, and demand — point forecasts with quantile bands; weather covariates are planned.
View caseTransportation: Road Speed & Congestion
Road-sensor speed and congestion series — per-segment point forecasts with quantile bands as a reference for staffing and capacity planning.
View caseMarket & Risk Series (planned)
Our next planned direction: market and risk series in finance — quantile envelopes that frame tail risk. Augments research, does not replace it, and is not investment advice.
View caseScope & Boundaries
What It Is — and Isn’t
It is:
- SequenxAI's general time-series foundation model, available on the SequenxAI Platform; its inputs are time series
- It is the model; data cleaning and quality reports come from the SequenxAI Platform
- Built for energy, transportation, finance and other time series, aiming to forecast new series directly
- Uncertainty expressed as quantile intervals that show the plausible range
- Delivery scope agreed in writing with each customer, per scenario
It is not:
- A general-purpose LLM or text / image / video generation model
- A model claimed to lead on accuracy without public same-scenario benchmarks
- A replacement for domain expert judgment
Getting started today
Want to forecaston your own data now?
The SequenxAI Platform provides forecasting through one API, cleaning data before it forecasts and returning auditable quality reports (beta). Contact the team when you want an evaluation on your own scenario.