ECMWF IFS vs ECMWF AIFS
ECMWF IFS or AIFS: is the AI model as good as the physics model?
Both come from ECMWF. IFS solves the equations of the atmosphere on a fine grid. AIFS is a machine-learning model trained on decades of past weather: it runs far faster and its large-scale forecasts are competitive with IFS, but its grid is coarser and it can smooth out small, intense features.
| ECMWF IFS | ECMWF AIFS | |
|---|---|---|
| Provider | ECMWF | ECMWF |
| Type | Global model | AI model |
| Area | Global | Global |
| Grid size | about 9 km | about 28 km |
| Updates | 4 times a day | 4 times a day |
| Looks ahead | 15 days | 15 days |
| Data licence | CC-BY-4.0 (open data) | CC-BY-4.0 (open data) |
Choose ECMWF IFS when
- You want fine detail and the physically based reference.
- You are looking at winds near coasts or around small, sharp features.
- You want the established model with the longest track record.
Choose ECMWF AIFS when
- You want a quick second opinion on the large-scale pattern.
- You want to see how an AI model reads the same situation.
- You are checking whether the models agree on a low-pressure track.
On the water
Treat AIFS as a second voice, not a replacement. If IFS and AIFS tell the same story, that is reassuring. If they differ, the difference is the interesting part.
More: AI weather models
Figures come from the Ilma catalog and the providers' own pages. Forecast skill changes as models are upgraded, so this page describes what each model is built for, not a ranking.