DreamerV3

Closed weights DeepMind,University of Toronto 200M parameters January 2023

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 Toronto
Organisation type
Industry,Academia
Country
United Kingdom of Great Britain and Northern Ireland, Canada
Published
10 January 2023
Authors
Danijar Hafner, Jurgis Pasukonis, Jimmy Ba, Timothy Lillicrap

What it does

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

Domain
Games
Task
Open ended play
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.

Parameters
200M

Table B1

Training data
1,600,000,000 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
2.2 × 10²⁰ FLOP

16 environment instances, each with 1 V100 running for 17 days (table A1) - it's not entirely clear if the GPU days already account for multiple environment instances. Assuming no: Compute: 17*24*60*60*125000000000000*0.3=5.508e+19 Assuming yes: Compute: 17*24*60*60*16*125000000000000*0.3=8.8128e+20 Geometric mean: 220320000000000000000

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
Chips used
16
Wall-clock time
6,528 hours (272 days)
Power draw
9.6 kW

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 https://github.com/danijar/dreamerv3

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

Using the same hyperparameters across all domains, DreamerV3 outperforms specialized model-free and model-based algorithms in a wide range of benchmarks and data-efficiency regimes. Applied out of the box, DreamerV3 also learns to obtain diamonds in the popular video game Minecraft from scratch given sparse rewards, a long-standing challenge in artificial intelligence for which previous approaches required human data or domain-specific heuristics.

Record confidence
Likely
Citations
1,073

Sources

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

Reference
Mastering Diverse Domains through World Models
Last updated
25 May 2026

What the numbers mean

Background

DreamerV3 was published by DeepMind,University of Toronto, in the country recorded as United Kingdom of Great Britain and Northern Ireland, during January 2023. The category the publisher falls under is industry,Academia.

It works in the domain of Games, and is recorded as performing the task of open ended play.

Its weights were never published, so it can only be reached through its provider. No graphics card changes that.

Training and provenance

Producing it required arithmetic totalling around 2.2 × 10²⁰ FLOP, on hardware recorded as NVIDIA V100. That figure measures what producing the model cost, and has no bearing on how fast it answers.

Training consumed a corpus of around 1,600,000,000 tokens of text.

It is tracked in the underlying dataset for one reason in particular: sOTA improvement.

Answers

DreamerV3 — common questions

01

DreamerV3— is it open source?

No. Its weights have not been published, so it exists only as a service controlled by its owner.

02

DreamerV3— how many parameters does it have?

It has a parameter count of 200M. Table B1. 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.

03

DreamerV3— who created it?

It was published by DeepMind,University of Toronto, based in United Kingdom of Great Britain and Northern Ireland, an organisation categorised as industry,Academia.

04

DreamerV3— when was it released?

It was published in January 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.

05

DreamerV3— what is it used for?

It works in the domain of Games, and is recorded as handling the task of open ended play. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

06

DreamerV3— how much compute was used to train it?

Training consumed around 2.2 × 10²⁰ FLOP, on hardware recorded as 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.

07

DreamerV3— what GPU do I need to run it?

None. This 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.

Source

Original publication

Record last updated 25 May 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.