One integration with the manufacturer's cloud API. Decisions reach the inverter through the same API. Nothing is installed on site. Trading appears in the manufacturer's own app, under its brand.
AI OS FOR INVERTERS
An automated AI energy trader for behind-the-meter batteries, built into the inverter. Trading becomes the default.
Introducing Farsees 2.0, our proprietary AI foundation model.
Backed and recognized by
Farsees 2.0
Farsees 2.0 turns idle battery capacity into trading assets.
Forecasting built for automated energy trading.
Farsees 2.0
High forecast accuracy.
Zero-shot forecasting.
Our proprietary AI foundation model
Farsees 2.0 forecasts site consumption, solar production and market prices. It is the core of the Solship Trading Operating System.
Published benchmarkEnergy market products
Farsees 2.0
Outperformed every published model tested.
- Google TimesFM, run zero-shot
- Amazon Chronos-2, run zero-shot
- Best public deep network, trained on the same features
Each product on its own held-out window, significance-tested. Solship white paper, Volume 01.
Probabilistic forecastForecast distribution
site historynow+37 h
37 h
forecast horizon
0 days
of site history needed
Zero-shot: a new site is forecast with 0 days of site history. The same model moves between sites without retraining.
White paper · Volume 01
Before the Gate · September 2026
The architecture, the benchmark and the results behind Farsees 2.0.
Volume 02 · comingEmbedded AI OS
Built into the inverter.
Solship activates every battery the inverter controls.
How it works
Inside the automated AI energy trader.
Solship Trading Operating SystemTrading · default
01Forecast
Farsees 2.0 forecasts site consumption, solar production and market prices.
02Millions of simulations
The most profitable path is selected across the full fleet.
03Dispatch
The inverter manufacturer’s full fleet follows the most profitable path.
Impact
Benefits for inverter manufacturers and battery owners.
€100K per MW per year
Gross revenue shared with inverter manufacturers and battery owners.
For inverter manufacturers
Recurring revenue
A shared revenue stream from the batteries controlled by the inverter.
Added value for customers
Automated energy trading built into the inverter.
For battery owners
Automated energy trading
Idle battery capacity becomes a trading asset.
Shared revenue
Battery owners participate in the value created by automated trading.
Team
The team behind the trader.
Inverter and battery system engineers, the machine learning team that built Farsees 2.0, and PhD advisors in optimization and energy markets.
Team experience from
Common questions.
Day-ahead, intraday and balancing. For each battery, every decision weighs the three markets against self-consumption.
Solship's proprietary AI foundation model for energy forecasting. It forecasts site consumption, solar production and market prices up to 37 hours ahead with 0 days of site history. The white paper reports the benchmark.
A new site is forecast with 0 days of site history. The same model moves between sites without retraining.
Yes. Trading can be switched off at any time, and the battery can be set to self-consumption or peak shaving. Every action on the battery is logged and reported.
Between the battery owner, the inverter manufacturer and Solship. Gross revenue is €100K per MW per year.