A faster view of the weather.
GraphCast produced skillful medium-range global weather forecasts using a learned model.
- Maturity
- Demonstrated, stage 3 of 4
- Support
- 1 source · paper
- Evidence detail
- How we know ↓
Weather forecasting demands substantial computation.
Numerical models simulate atmospheric physics.
A graph neural network learned weather evolution from historical data.
What was shown, and what wasn’t
Shown
The reported evaluation beat the operational comparison on more than 90% of 1,380 targets.
Not shown · limits
- Retrospective benchmark skill does not establish local warning quality.
Bright editorial interpretation
Faster forecasts could help forecasters explore conditions sooner.
Still open
Evaluate extremes and operational integration.
How we know1 source · checked 2026-09-05 · no corrections
Original sources
Institutions: Google DeepMind
- Maturity
- Demonstrated
- Source published
- 2023-11-14
- Last source review
- 2026-09-05
- Editorial method
- Original source check
- Place / relevance
- London, United Kingdom · institution-location
Bright compared this account with the linked original and supporting sources and kept reported, budgeted, projected, and observed claims distinct. Bright did not independently audit the underlying records.
Maturity describes the tested or operational setting. Confidence describes support for the particular claim; one does not determine the other.
Revision & correction history
No corrections recorded.
Keep exploring
Explore the shared question in another setting. These connections do not imply replication.