JEST++

Closed weights Google DeepMind June 2024

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
Google DeepMind
Organisation type
Industry
Country
United States of America
Published
25 June 2024
Authors
Talfan Evans, Nikhil Parthasarathy, Hamza Merzic, Olivier J. Henaff

What it does

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

Domain
Vision
Task
Image classification

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
131,072,000,000,000 tokens

4b training examples seen (figure 1) The vision encoder takes images resized to (256 x 256) and the text-encoder tokenizes text with the sentencepiece tokenizer [26] trained on the English C4 dataset [39]. We crop the text to the first 64 tokens. visual encoder: ViT-B/16 4*10^9*((256/16)^2 + 64) = 1.28e+12 tokens (text and image)

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
1.9 × 10²¹ FLOP

1.9*10^21 from figure 1 and table 1 there is slight disrepancy (within confidence interval) between sigclip/jest++ training compute ratio if we compare epoch's estimated compute and table 1

How it was established
Other

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
Google TPU v5e
Chips used
256
Power draw
113.7 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
Unreleased

How it is classified

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

Record confidence
Confident
Citations
30

Sources

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

Reference
Data curation via joint example selection further accelerates multimodal learning
Last updated
25 May 2026

What the numbers mean

Background

JEST++ was published by Google DeepMind, in the country recorded as United States of America, during June 2024. It comes out of an organisation categorised as industry.

It works in the domain of Vision, and is recorded as performing the task of image classification.

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

Training and provenance

Producing it required arithmetic totalling around 1.9 × 10²¹ FLOP, on hardware recorded as Google TPU v5e. That figure measures what producing the model cost, and has no bearing on how fast it answers.

It was trained on a corpus of about 131,072,000,000,000 tokens of text.

Answers

JEST++ — common questions

01

JEST++— is it open source?

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

02

JEST++— how many parameters does it have?

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

03

JEST++— who created it?

It was published by Google DeepMind, based in United States of America, an organisation categorised as industry.

04

JEST++— when was it released?

It was published in June 2024. 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

JEST++— what is it used for?

It works in the domain of Vision, and is recorded as handling the task of image classification. 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

JEST++— how much compute was used to train it?

Training consumed around 1.9 × 10²¹ FLOP, on hardware recorded as Google TPU v5e. 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

JEST++— 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.