Validation
Our engine, validated against industry references
During the beta, the only public proof is our validation reports: a concrete number per report, traceable methodology and downloadable PDFs.
Reports
The three reports
- Rear-irradiance bias +1.50% (annual, SAT, one site)
Validation report — NREL BEST bifacial
Validation of the bifacial model against the NREL BEST reference.
Download PDF - Agreement within 2%
Validation report — Sandia / PVPMC
Validation against the Sandia / PVPMC reference models.
Download PDF Read in the browser - Deviation 0.75%
Production benchmark — JUST 5 MWp case
Model compared against real measured production at a 5 MWp plant.
Download PDF
Limits
What these results do not show
- Validated against measured production at the NIST Ground Array: −1.1% on annual energy against 371 MWh measured. One plant and one period. That does not validate every model, climate or plant type. We make no claim of being more accurate than any other tool.
- Energy yield and financial returns are estimates based on your project assumptions. Task automation does not replace engineering review.
Next step
Protocol, result and limits — published
Every report states what was compared, against what, with what result and within what limits. If you want to see the engine run on your own site rather than read about it, talk to an engineer.
Request a session with an engineer