Model Tracker
V-JEPA
V-JEPA is an AI model made by Meta. Meta released it on 15 February 2024. Its weights are open.
- Released
- 15 February 2024 — ai.meta.com, read 16 October 2026
- Status
- Not confirmed: The maker's pages reviewed do not assign V-JEPA one of the register's status labels or establish that it is currently active, in preview, deprecated, or retired.
- Takes
- Video — ai.meta.com, read 16 October 2026
- Returns
- Not confirmed: The maker describes predictions in an abstract representation space, but does not identify a returned output as one of the register's listed modalities.
- Open weights
- Yes — github.com, read 16 October 2026
- Licence
- Not confirmed: two readings disagree: Attribution-NonCommercial 4.0 International and Creative Commons NonCommercial
- Context window
- Not confirmed: The maker's pages reviewed do not print a context-window size for V-JEPA.
- Longest answer
- Not confirmed: The maker's pages reviewed do not print a standard maximum output-token limit for V-JEPA.
- Knowledge cutoff
- Not confirmed: The maker's pages reviewed do not state a knowledge-cutoff date for V-JEPA.
- API price, input
- Not confirmed: The maker's pages reviewed do not publish a standard API input price for V-JEPA.
- API price, output
- Not confirmed: The maker's pages reviewed do not publish a standard API output price for V-JEPA.
- Where
- Downloadable weights — github.com, read 16 October 2026
- Parameters
- Not confirmed: The maker publishes model variants such as ViT-L and ViT-H but does not print their parameter counts on the pages reviewed.
- Model card
- Not confirmed: The maker's pages reviewed link to a research paper and code repository, but do not provide a model card.
Questions
When was V-JEPA released?
Meta released V-JEPA on 15 February 2024.
Are the weights of V-JEPA open?
Yes: Meta publishes the weights of V-JEPA for download.
More from Meta
- Audiobox
- AudioGen
- Chameleon
- CM3leon
- Code Llama
- DINOv2
- DINOv3
- Emu
- Emu Edit
- Emu Video
- EnCodec
- I-JEPA
- ImageBind
- LLaMA
- Llama 2
- Llama 3
- Llama 3.1
- Llama 3.2
- Llama 3.3
- Llama 4 Maverick
- Llama 4 Scout
- M2M-100
- Make-A-Video
- Movie Gen
Sources
- V-JEPA: The next step toward Yann LeCun’s vision of advanced machine intelligence (AMI) (read 16 October 2026)
- GitHub - facebookresearch/jepa: PyTorch code and models for V-JEPA self-supervised learning from video. (read 16 October 2026)
- jepa/LICENSE at main · facebookresearch/jepa · GitHub (read 16 October 2026)