HRA

Closed weights Maluuba,Microsoft June 2017

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
Maluuba,Microsoft
Organisation type
Industry,Industry
Country
Canada, United States of America
Published
13 June 2017
Authors
H Van Seijen, M Fatemi, J Romoff, R Laroche

What it does

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

Domain
Games
Task
Atari game: Ms. Pac-Man
Numerical format
FP32

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
150,000,000 tokens

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

Table 1 "With the best combination, HRA not only outperforms the state-of-the-art on both metrics, it also significantly outperforms the human score, convincingly demonstrating the strength of HRA."

Record confidence
Unknown
Citations
285

Sources

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

Reference
Hybrid Reward Architecture for Reinforcement Learning
Last updated
25 May 2026

What the numbers mean

What this model is

HRA was published by Maluuba,Microsoft, in Canada, in June 2017. industry,Industry is the category the publisher falls under.

It works in Games, and is recorded as doing atari game: Ms. Pac-Man.

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

How it was trained

The training set ran to roughly 150,000,000 tokens.

The reason it appears in this catalogue at all is sOTA improvement.

Answers

HRA — common questions

01

Who created HRA?

HRA was published by Maluuba,Microsoft, based in Canada, categorised as industry,Industry.

02

When was HRA released?

HRA was published in June 2017. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

03

What is HRA used for?

HRA works in Games, and is recorded as handling atari game: Ms. Pac-Man. These are the areas it was designed around; they describe intent rather than a hard boundary.

04

What GPU do I need to run HRA?

None. HRA 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.

05

Is HRA open source?

The licensing for HRA was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

06

How many parameters does HRA have?

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

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.