EfficientZero

Closed weights Tsinghua University,University of California (UC) Berkeley,Shanghai Qi Zhi institute October 2021

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
Tsinghua University,University of California (UC) Berkeley,Shanghai Qi Zhi institute
Organisation type
Academia,Academia
Country
China, United States of America
Published
30 October 2021
Authors
Weirui Ye, Shaohuai Liu, Thanard Kurutach, Pieter Abbeel, Yang Gao

What it does

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

Domain
Games
Task
Atari
Numerical format
FP16

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
100,000 tokens

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

GPL-3.0 (copyleft + prohibits incorporating the software into proprietary software) https://github.com/YeWR/EfficientZero?tab=readme-ov-file

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

"Our method is 176% and 163% better than the previous SoTA performance, in mean and median human normalized score respectively" Table 1 "EfficientZero achieves superhuman performance with only 2 hours of real-time game play. Our method is 176% and 163% better than the previous SoTA performance, in mean and median human normalized score respectively." Table 2: Scores achieved by EfficientZero (mean & standard deviation for 10 seeds) and some baselines on some low-dimensional environments on the…

Record confidence
Unknown
Citations
324

Sources

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

Reference
Mastering Atari Games with Limited Data
Last updated
25 May 2026

What the numbers mean

What this model is

EfficientZero was published by Tsinghua University,University of California (UC) Berkeley,Shanghai Qi Zhi institute, in China, in October 2021. It comes out of academia,Academia.

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

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

What went into building it

Around 100,000 tokens went into training it.

Its inclusion criterion is sOTA improvement.

Answers

EfficientZero — common questions

01

Is EfficientZero open source?

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

02

How many parameters does EfficientZero have?

No parameter count has been published for EfficientZero, which is why no memory or speed figure appears on this page.

03

Who created EfficientZero?

EfficientZero was published by Tsinghua University,University of California (UC) Berkeley,Shanghai Qi Zhi institute, based in China, categorised as academia,Academia.

04

When was EfficientZero released?

EfficientZero was published in October 2021. 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

What is EfficientZero used for?

EfficientZero works in Games, and is recorded as handling atari. 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.

06

What GPU do I need to run EfficientZero?

None. EfficientZero 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.