Search public pages, research tools, and SAVRN solutions.
Open-weight model · Video classification
by AI at Meta facebook/vjepa2-vitg-fpc32-384-diving48
A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of VJEPA, resulting in state-of-the-art video understanding capabilities, leveraging data and model sizes at scale. The code is released in this repository.
What it takes to serve vjepa2-vitg-fpc32-384-diving48 (1.1B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 2.3 GB | 2.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 1.1 GB | 1.4 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.6 GB | 0.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.
By AI at Meta, published under mit, revision 0b4824337531.
A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of VJEPA, resulting in state-of-the-art video understanding capabilities, leveraging data and model sizes at scale. The code is released in this repository. This is V-JEPA 2 ViT-g 384 model with video classification head pretrained on Diving 48 dataset. To run V-JEPA 2 model, ensure you have installed the latest transformers
A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of VJEPA, resulting in state-of-the-art video understanding capabilities, leveraging data and model sizes at scale. The code is released in this repository.
To run V-JEPA 2 model, ensure you have installed the latest transformers:
pip install -U git+https://github.com/huggingface/transformers
import torch
import numpy as np
from torchcodec.decoders import VideoDecoder
from transformers import AutoVideoProcessor, AutoModelForVideoClassification
device = "cuda" if torch.cuda.is_available() else "cpu"
# Load model and video preprocessor
hf_repo = "facebook/vjepa2-vitg-fpc32-384-diving48"
model = AutoModelForVideoClassification.from_pretrained(hf_repo).to(device)
processor = AutoVideoProcessor.from_pretrained(hf_repo)
# To load a video, sample the number of frames according to the model.
video_url = "https://huggingface.co/facebook/vjepa2-vitg-fpc32-384-diving48/resolve/main/sample/diving.mp4"
vr = VideoDecoder(video_url)
frame_idx = np.arange(0, model.config.frames_per_clip, 8) # you can define more complex sampling strategy
video = vr.get_frames_at(indices=frame_idx).data # frames x channels x height x width
# Preprocess and run inference
inputs = processor(video, return_tensors="pt").to(model.device)
with torch.no_grad():
outputs = model(**inputs)
logits = outputs.logits
print("Top 5 predicted class names:")
top5_indices = logits.topk(5).indices[0]
top5_probs = torch.softmax(logits, dim=-1).topk(5).values[0]
for idx, prob in zip(top5_indices, top5_probs):
text_label = model.config.id2label[idx.item()]
print(f" - {text_label}: {prob:.2f}")
Output:
Top 5 predicted class names:
- ['Forward', '35som', 'NoTwis', 'PIKE']: 0.49
- ['Forward', '25som', 'NoTwis', 'PIKE']: 0.13
- ['Forward', '25som', '1Twis', 'PIKE']: 0.13
- ['Forward', '35som', 'NoTwis', 'TUCK']: 0.10
- ['Forward', '25som', '2Twis', 'PIKE']: 0.04
@techreport{assran2025vjepa2,
title={V-JEPA~2: Self-Supervised Video Models Enable Understanding, Prediction and Planning},
author={Assran, Mahmoud and Bardes, Adrien and Fan, David and Garrido, Quentin and Howes, Russell and
Komeili, Mojtaba and Muckley, Matthew and Rizvi, Ammar and Roberts, Claire and Sinha, Koustuv and Zholus, Artem and
Arnaud, Sergio and Gejji, Abha and Martin, Ada and Robert Hogan, Francois and Dugas, Daniel and
Bojanowski, Piotr and Khalidov, Vasil and Labatut, Patrick and Massa, Francisco and Szafraniec, Marc and
Krishnakumar, Kapil and Li, Yong and Ma, Xiaodong and Chandar, Sarath and Meier, Franziska and LeCun, Yann and
Rabbat, Michael and Ballas, Nicolas},
institution={FAIR at Meta},
year={2025}
}
7 files, 4.5 GB in total. The weights are 1 file totalling 4.5 GB in safetensors.
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 4.5 GB | ec1831cda77a |
| config.json | Configuration | 4.2 KB | — |
| video_preprocessor_config.json | Configuration | 1.5 KB | — |
| README.md | Documentation | 3.7 KB | — |
| notebook_finetuning.ipynb | Other | 544.8 KB | — |
| sample/diving.mp4 | Other | 324.1 KB | 7652d0c08ebe |
| .gitattributes | Repository | 1.6 KB | — |
Released by AI at Meta through its official repository on Hugging Face. Read the license.
| Precision | Weights in memory |
|---|---|
| As published | 4.5 GB |
| 16-bit | 2.3 GB |
| 8-bit | 1.1 GB |
| 4-bit | 0.6 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
About 2.7 GB at 16-bit and 0.7 GB at 4-bit: the weights (1.1B parameters) plus a working margin. A long context needs more.
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Yes. vjepa2-vitg-fpc32-384-diving48 is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.
A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of VJEPA, resulting in state-of-the-art video understanding capabilities, leveraging data and model sizes at scale. The code is released in this repository. This is V-JEPA 2 ViT-g 384 model with video classification head pretrained on Something-Something-V2 dataset. To run V-JEPA 2 model, ensure you have installed the latest transformers
A HuggingFace-format conversion of Meta AI's V-JEPA 2.1 ViT-g/16 video encoder and predictor, operating at 384x384 resolution. The weights are Meta's, copied without modification. This repository provides the transformers-compatible packaging plus a documented numerical validation against the original implementation. No prior HuggingFace port of this variant existed at the time of upload. The only structural change is that the fused QKV projection of each attention block is split into separate query / key / value matrices, following the convention used by transformers. This is a re-parameterization, not a change of weights. It is also convenient downstream: PEFT adapters apply to…
Cosmos-Embed1 is a joint video-text embedder tailored for physical AI. It can be used for text-to-video retrieval, inverse video search, semantic deduplication, zero-shot and k-nearest-neighbors (kNN) classification, and as a base model for video curation tasks. It has state-of-the-art (SOTA) performance on autonomous vehicle (AV) and robotics datasets, while maintaining competitive performance in general domains. A fine-tuned variant is also provided for video anomaly detection and classification. This model is ready for commercial use. The Cosmos-Embed1 release includes the following embedders: Note: while each checkpoint was optimized at a specific fixed resolution (and default to…
A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of VJEPA, resulting in state-of-the-art video understanding capabilities, leveraging data and model sizes at scale. The code is released in this repository. To run V-JEPA 2 model, ensure you have installed the latest transformers: V-JEPA 2 is intended to represent any video (and image) to perform video classification, retrieval, or as a video encoder for VLMs. To load a video, sample the number of frames according to the model. For this model, we use 64. To load an image, simply copy the image to the desired number of frames. For more code examples, please refer to the V-JEPA 2…
A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of VJEPA, resulting in state-of-the-art video understanding capabilities, leveraging data and model sizes at scale. The code is released in this repository. To run V-JEPA 2 model, ensure you have installed the latest transformers: V-JEPA 2 is intended to represent any video (and image) to perform video classification, retrieval, or as a video encoder for VLMs. To load a video, sample the number of frames according to the model. For this model, we use 64. To load an image, simply copy the image to the desired number of frames. For more code examples, please refer to the V-JEPA 2…
VideoMAEv2-giant model pre-trained for 1200 epochs in a self-supervised way on UnlabeldHybrid-1M dataset. It was introduced in the paper [[CVPR23]VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking](https://arxiv.org/abs/2203.12602) by Wang et al. and first released in GitHub. You can use the raw model for video feature extraction. Here is how to use this model to extract a video feature