GraphCast
No estimate
No hardware requirements for this model
This model's weights are open, but no parameter count has been published for it. Every memory and speed figure starts from that number, so we would rather show nothing than a fabricated estimate.
On record
Full specification
Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.
Origin
Who built this model, where, and when it was published.
- Organisation
- Google DeepMind
- Organisation type
- Industry
- Country
- United States of America
- Published
- 14 November 2023
- Authors
- Remi Lam, Alvaro Sanchez-Gonzalez, Matthew Willson, Peter Wirnsberger, Meire Fortunato, Ferran Alet, Suman Ravuri, Timo Ewalds, Zach Eaton-Rosen, Weihua Hu, Alexander Merose, Stephan Hoyer, George Holland, Oriol Vinyals, Jacklynn Stott, Alexander Pritzel, Shakir Mohamed, Peter Battaglia
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Earth science
- Task
- Weather forecasting
- Numerical format
- BF16
Size
How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.
- Training data
- tokens
Training compute
The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.
- Training compute
- 2.1 × 10²² FLOP
- How it was established
- Hardware
"Training GraphCast took roughly four weeks on 32 Cloud TPU v4 devices using batch parallelism." 4.6: "we use bfloat16 floating point precision" 2.1e22 = 2.75E+14 FLOP/s * 32 * 60* 60 * 24 * 7 * 4
Availability
Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.
- Weights
- Open — downloadable
- Model access
- Open weights (non-commercial)
- Training code
- Open source
Apache 2 for training and inference code https://github.com/google-deepmind/graphcast CC BY-NC-SA 4.0 for weights
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Why it is tracked
- SOTA improvement
- Record confidence
- Speculative
"Our state-of-the-art model delivers 10-day weather predictions at unprecedented accuracy in under one minute"
Sources
Where this record came from and when it was last checked.
- Reference
- Learning skillful medium-range globalweather forecasting
- Last updated
- 28 November 2025
What the numbers mean
Background
GraphCast was published by Google DeepMind, in United States of America, in November 2023. It comes out of industry.
It works in Earth science, and is recorded as doing weather forecasting.
Because its weights were released, nothing about running it depends on a provider staying available — it is yours once downloaded.
How it was trained
Training it took roughly 2.1 × 10²² FLOP of computation — a measure of what producing the model cost, not of how fast it answers.
It is tracked in the underlying dataset for one reason in particular: sOTA improvement.
Answers
GraphCast — common questions
Who created GraphCast?
GraphCast was published by Google DeepMind, based in United States of America, categorised as industry.
When was GraphCast released?
GraphCast was published in November 2023. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
What is GraphCast used for?
GraphCast works in Earth science, and is recorded as handling weather forecasting. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.
Where can I download GraphCast?
The weights for GraphCast are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.
How much compute was used to train GraphCast?
Around 2.1 × 10²² FLOP. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.
What GPU do I need to run GraphCast?
We cannot say. GraphCast has open weights, but no parameter count has been published for it, and every memory and speed calculation starts from that number. We would rather show nothing than a fabricated estimate.
Is GraphCast open source?
Its weights are published, so GraphCast can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.
How many parameters does GraphCast have?
No parameter count has been published for GraphCast, which is why no memory or speed figure appears on this page.
The other direction
Looking at it from the other side?
This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.