V-JEPA
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
- Meta AI
- Organisation type
- Industry
- Country
- United States of America
- Published
- 15 February 2024
- Authors
- Adrien Bardes, Quentin Garrido, Jean Ponce, Xinlei Chen, Michael Rabbat, Yann LeCun, Mahmoud Assran, Nicolas Ballas
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Vision
- Task
- Representation 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.
- Parameters
- 630M
- Training data
- tokens
They train 200M and 630M models (see e.g. table 6)
2900k video clips (Table 2)
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
- 1.6 × 10²¹ FLOP
Figure 7 shows that each example (16 frames of 224*224 resoultion) is represented with 1568 tokens. Table 8 specifies (for ViT-H/16_224): Batch size = 3072 Iterations = 90000 Total training tokens: 90000 iterations *3072 batch size * 1568 tokens =433520640000 Training compute using 6ND (ignoring fine-tuning): 6*630000000*433520640000 = 1638708019200000000000
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 A100
- Wall-clock time
- 50 hours
Figure 5
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Confident
Sources
Where this record came from and when it was last checked.
- Reference
- Revisiting Feature Prediction for Learning Visual Representations from Video
- Last updated
- 28 November 2025
What the numbers mean
Where it came from
V-JEPA was published by Meta AI, in the country recorded as United States of America, during February 2024. The category the publisher falls under is industry.
It works in the domain of Vision, and is recorded as performing the task of representation learning.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
How it was trained
The training run consumed about 1.6 × 10²¹ FLOP, on hardware recorded as NVIDIA A100. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Answers
V-JEPA — common questions
V-JEPA— how many parameters does it have?
It has a parameter count of 630M. They train 200M and 630M models (see e.g. table 6). 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.
V-JEPA— who created it?
It was published by Meta AI, based in United States of America, an organisation categorised as industry.
V-JEPA— when was it released?
It was published in February 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.
V-JEPA— what is it used for?
It works in the domain of Vision, and is recorded as handling the task of representation learning. 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.
V-JEPA— how much compute was used to train it?
Training consumed around 1.6 × 10²¹ FLOP, on hardware recorded as NVIDIA A100. 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.
V-JEPA— 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.
V-JEPA— is it open source?
The licensing was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.
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.