ASE

Closed weights NVIDIA,University of California (UC) Berkeley May 2022

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
NVIDIA,University of California (UC) Berkeley
Organisation type
Industry,Academia
Country
United States of America
Published
5 May 2022
Authors
Xue Bin Peng, Yunrong Guo, Lina Halper, Sergey Levine, Sanja Fidler

What it does

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

Domain
Robotics
Task
Animal (human/non-human) imitation
Approach
Reinforcement learning

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
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
4.5 × 10¹⁹ FLOP

Training was done using the Isaac Gym simulator on an NVIDIA V100 GPU. The model was trained on over 10 billion samples, which equates to 10 years of simulated experience time. Training took around 10 days on a single GPU. 1.3e14 * 10 * 24 * 3600 * 0.4 = 4.49e19

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 Tesla V100 PCIe 16 GB
Wall-clock time
240 hours (10 days)

Training took around 10 days on a single GPU.

How it is classified

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

Record confidence
Likely
Citations
56

Sources

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

Reference
ASE: Large-Scale Reusable Adversarial Skill Embeddings for Physically Simulated Characters
Last updated
28 November 2025

What the numbers mean

About this model

ASE was published by NVIDIA,University of California (UC) Berkeley, in United States of America, in May 2022. The organisation is categorised as industry,Academia.

It works in Robotics, and is recorded as doing animal (human/non-human) imitation.

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

What went into building it

The training run consumed about 4.5 × 10¹⁹ FLOP, on NVIDIA Tesla V100 PCIe 16 GB. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

Answers

ASE — common questions

01

Is ASE open source?

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

02

How many parameters does ASE have?

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

03

Who created ASE?

ASE was published by NVIDIA,University of California (UC) Berkeley, based in United States of America, categorised as industry,Academia.

04

When was ASE released?

ASE was published in May 2022. 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 ASE used for?

ASE works in Robotics, and is recorded as handling animal (human/non-human) imitation. 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

How much compute was used to train ASE?

Around 4.5 × 10¹⁹ FLOP, on NVIDIA Tesla V100 PCIe 16 GB. 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

What GPU do I need to run ASE?

None. ASE 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 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.