Hanabi 4 player

Closed weights DeepMind,University of Oxford,Carnegie Mellon University (CMU),Google Brain 764K parameters February 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,University of Oxford,Carnegie Mellon University (CMU),Google Brain
Organisation type
Industry,Academia,Academia,Industry
Country
United Kingdom of Great Britain and Northern Ireland, United States of America
Published
1 February 2019

What it does

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

Domain
Games
Task
Hanabi

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
764K

source: https://docs.google.com/spreadsheets/d/1Kj4Q5WADcDXtUJLIOfGTCE3tGvxNczEMwyy8QtgSkHk/edit#gid=54587040&fvid=1361937389

Training data
20,000,000,000 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
4.3 × 10¹⁸ FLOP

14.13e+12 FLOP/s * 7 days * 86400 s/day * 0.50 utilization = 4.3e+18 FLOP

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
NVIDIA V100

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
Unreleased

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
Historical significance

Adapted some SOTA RL algorithms to a new task that posed research challenges

Record confidence
Confident
Citations
229

Sources

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

Reference
The Hanabi Challenge: A New Frontier for AI Research
Last updated
28 November 2025

What the numbers mean

About this model

Hanabi 4 player was published by DeepMind,University of Oxford,Carnegie Mellon University (CMU),Google Brain, in United Kingdom of Great Britain and Northern Ireland, in February 2019. industry,Academia,Academia,Industry is the category the publisher falls under.

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

Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.

What went into building it

The training run consumed about 4.3 × 10¹⁸ FLOP, on NVIDIA V100. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

Around 20,000,000,000 tokens went into training it.

The reason it appears in this catalogue at all is historical significance.

Answers

Hanabi 4 player — common questions

01

How much compute was used to train Hanabi 4 player?

Around 4.3 × 10¹⁸ FLOP, on NVIDIA V100. 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.

02

What GPU do I need to run Hanabi 4 player?

None. Hanabi 4 player 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.

03

Is Hanabi 4 player open source?

No. Hanabi 4 player has not had its weights published, so it exists only as a service controlled by its owner.

04

How many parameters does Hanabi 4 player have?

Hanabi 4 player has 764K parameters. source: https://docs.google.com/spreadsheets/d/1Kj4Q5WADcDXtUJLIOfGTCE3tGvxNczEMwyy8QtgSkHk/edit#gid=54587040&fvid=1361937389. 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.

05

Who created Hanabi 4 player?

Hanabi 4 player was published by DeepMind,University of Oxford,Carnegie Mellon University (CMU),Google Brain, based in United Kingdom of Great Britain and Northern Ireland, categorised as industry,Academia,Academia,Industry.

06

When was Hanabi 4 player released?

Hanabi 4 player was published in February 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.

07

What is Hanabi 4 player used for?

Hanabi 4 player works in Games, and is recorded as handling hanabi. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

Source

Original publication

Record last updated 28 November 2025

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.