Design an energy system. Optimise it properly.
Levelised is a techno-economic optimiser that runs in your browser. Drag generators, storage, loads and links onto a canvas (electricity, heat, hydrogen, water, any energy or mass flow) and solve a full year at hourly resolution with a real optimisation engine, built on the open-source PyPSA framework.
Energy decisions are core business now
Max van Someren, founder of Levelised, on why we built it and how it works.
Describe your site and the assistant builds the model
Placing every component, finding a cost for each one and loading a trace for every renewable is fine for the final model and tedious for the first five. The assistant in the app does that first pass from a description. It lays out the buses, the load and the supply options on your canvas, takes technology costs from CSIRO GenCost and emission factors from the National Greenhouse Accounts Factors, builds solar and wind traces for the location, and solves the year when you ask it to.
Every figure it sets carries its source, so you can click a component and see where the number came from. Reference costs are screening figures and the assistant can get things wrong, so check the inputs before you rely on the answer.
It comes with a weekly allowance, shown in the assistant window, that resets every Monday. The paid tier has ten times the allowance.
Things people ask it
- A 100 MW mine in the Pilbara, off-grid. What is the cheapest mix of solar, wind, batteries and gas?
- Does adding solar bring the cost of energy down at this site, and how much should I build?
- Why did my solve come back infeasible?
Built for the decisions in your sector
-
Data centres
Time-to-power, hourly carbon-free energy commitments, and what a grid, on-site or hybrid power strategy actually costs per MWh.
Explore data centres → -
Mining & metals
Off-grid and remote power, diesel displacement, hybrid plant design, and process heat and water in the same optimisation.
Explore mining & metals → -
Renewable energy
Storage sizing, hybrid plant design, co-location with large loads, and what a plant can credibly commit to under a PPA.
Explore renewable energy → -
Renewable fuels
Green hydrogen, ammonia and methanol chains: electrolyser utilisation, storage buffers, and the real cost per tonne.
Explore renewable fuels →
How it works
-
Build
Lay out your system on the canvas: generators, storage, loads and the links between them. Any carrier (power, heat, hydrogen, water) in one connected model.
-
Solve
Optimise capacity and dispatch over 8,760 hourly snapshots, a full year, using the open-source PyPSA framework and the HiGHS solver. Least-cost sizing and operation, not a spreadsheet approximation.
-
Download
Save your work to the cloud and download your results as CSV on the Demo tier. The paid tier also downloads the project file and a PyPSA netCDF, so your model is never locked in.
Start on the Demo tier. More for teams that need it.
Sign in with your work email and the Demo tier is yours at no cost: build, solve, save and download, with the assistant to help. The advanced layers below are the paid tier (see what the paid features do), and if they matter to you, tell us and we'll be in touch.
| Capability | Demo | Paid |
|---|---|---|
| Canvas modelling: generators, storage, loads, links, any carrier | Included | Included |
| Hourly optimisation over a full year (8,760 snapshots) | Included | Included |
| AI assistant: builds, edits and explains models from a conversation | Weekly allowance | Ten times the allowance |
| Components per model | Up to 20 | Unlimited |
| Snapshots per solve | 8,760 (one hourly year) | Unlimited (multi-period) |
| Solves per day | 10 | Unlimited |
| Solve time limit | 5 minutes | Unlimited |
| Compute | Standard | Enhanced |
| Downloads | Results CSV | Results CSV, project file, PyPSA netCDF |
| Imports | Time-series CSVs | Time-series CSVs, PyPSA netCDF network files |
| Cloud saves of solved results | Statistics level | Full hourly time series |
| Scenario management | — | Included |
| Multi-period analysis | — | Included |
| Custom optimisation constraints and objectives | — | Included |
| Binding limits: which limits shaped the solve | — | Included |
| Connect your own AI (MCP) | — | Included |