AlphaStar

Closed weights DeepMind 139M parameters October 2019

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

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
DeepMind
Organisation type
Industry
Country
United Kingdom of Great Britain and Northern Ireland
Published
30 October 2019
Authors
Oriol Vinyals,Igor Babuschkin,Wojciech M. Czarnecki,Michaël Mathieu,Andrew Dudzik,Junyoung Chung,David H. Choi,Richard Powell,Timo Ewalds,Petko Georgiev,Junhyuk Oh,Dan Horgan,Manuel Kroiss,Ivo Danihelka,Aja Huang,Laurent Sifre,Trevor Cai,John P. Agapiou,Max Jaderberg,Alexander S. Vezhnevets,Rémi Leblond,Tobias Pohlen,Valentin Dalibard,David Budden,Yury Sulsky,James Molloy,Tom L. Paine,Caglar Gulce…

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Games
Task
StarCraft

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.

Parameters
139M

AlphaStar has 139 million weights, but only 55 million weights are required during inference.

Training data
tokens

Multiple data types. First supervised learning, then other stuff

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
1.1 × 10²³ FLOP

(Estimate by James Sanders, checked by Robi Rahman) Fig 6 indicates that the learner uses 16 TPU "devices" which are 128 TPU cores total, which matches 4 TPUs per device, and 2 cores per TPU. Fig 6 indicates that 64 TPUs are used for training, and 64 are used for inference. (128 TPUs)*(12 agents)*(44 days)*(123 TFLOPS)*(0.3 utilization) = 2.15e23 FLOP (total), of which 1.07e23 FLOP are for training.

How it was established
Hardware

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware
Google TPU v3
Chips used
384
Chip-hours
405,504
Wall-clock time
1,056 hours (44 days)

"Each agent was trained using 32 third-generation tensor processing units (TPUs) over 44 days"

Power draw
354.2 kW
Compute cost
$130,654

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
Closed — provider access only
Model access
Unreleased
Training code
Open source

Apache 2.0, training tools: https://github.com/google-deepmind/alphastar training instructions here: https://github.com/google-deepmind/alphastar/blob/main/alphastar/unplugged/README.md

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Frontier model
Yes
Why it is tracked
Discretionary
Record confidence
Confident
Citations
4,093

Sources

Where this record came from and when it was last checked.

Reference
Grandmaster level in StarCraft II using multi-agent reinforcement learning
Last updated
1 January 2026

What the numbers mean

Background

AlphaStar was published by DeepMind, in United Kingdom of Great Britain and Northern Ireland, in October 2019. It comes out of industry.

It works in Games, and is recorded as doing starCraft.

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

How it was trained

The training run consumed about 1.1 × 10²³ FLOP, on Google TPU v3. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

Its inclusion criterion is discretionary.

Answers

AlphaStar — common questions

01

What is AlphaStar used for?

AlphaStar works in Games, and is recorded as handling starCraft. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

02

How much compute was used to train AlphaStar?

Around 1.1 × 10²³ FLOP, on Google TPU v3. 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.

03

What GPU do I need to run AlphaStar?

None. AlphaStar is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

04

Is AlphaStar open source?

No. AlphaStar has not had its weights published, so it exists only as a service controlled by its owner.

05

How many parameters does AlphaStar have?

AlphaStar has 139M parameters. AlphaStar has 139 million weights, but only 55 million weights are required during inference. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

06

Who created AlphaStar?

AlphaStar was published by DeepMind, based in United Kingdom of Great Britain and Northern Ireland, categorised as industry.

07

When was AlphaStar released?

AlphaStar was published in October 2019. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

Source

Original publication

Record last updated 1 January 2026

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.