# videomae-large by Multimedia Computing Group-Nanjing University Source: https://savrn.com/models/videomae-large Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve videomae-large (343M 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 | 0.7 GB | 0.8 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 0.3 GB | 0.4 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 0.2 GB | 0.2 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/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](https://savrn.com/ai-index/pricing/gpus), read Oct 24, 2026. [videomae-large on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/videomae-large/gpus) ## Model Card VideoMAE model pre-trained on Kinetics-400 for 1600 epochs in a self-supervised way. It was introduced in the paper VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training by Tong et al. and first released in this repository. Disclaimer: The team releasing VideoMAE did not write a model card for this model so this model card has been written by the Hugging Face team. VideoMAE is an extension of Masked Autoencoders (MAE) to video. The architecture of the model is very similar to that of a standard Vision Transformer (ViT), with a decoder on top for predicting pixel values for masked patches. Videos are presented to the model as a sequence of… Excerpt from the card by Multimedia Computing Group-Nanjing University, licensed cc-by-nc-4.0. ## Configuration Architecture VideoMAEForPreTraining Layers 24 Hidden size 1,024 Feed-forward size 4,096 Attention heads 16 Stored precision float32 Model type videomae ## Identity and Version Repository MCG-NJU/videomae-large Publisher Multimedia Computing Group-Nanjing University Task Video classification Modality Video Library transformers Parameters 343M parameters Languages Not stated by the source Revision 12da269a02d3e1fbbb7011e610e591fea8061dca First published 2022-08-02 Last updated 2024-04-01 ## Files and Weights 6 files, 2.7 GB in total. The weights are 2 files totalling 2.7 GB in bin, safetensors. Weights2 files · 2.7 GB Configuration2 files · 998 B Documentation1 file · 3.8 KB Repository1 file · 1.4 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 1.4 GB | 243706355bfd | | pytorch_model.bin | Weights | 1.4 GB | c7549bea23c2 | | config.json | Configuration | 727 B | — | | preprocessor_config.json | Configuration | 271 B | — | | README.md | Documentation | 3.8 KB | — | | .gitattributes | Repository | 1.4 KB | — | ## License and Download License cc-by-nc-4.0 Access Open weights, no gate Download size 2.7 GB [Download from Multimedia Computing Group-Nanjing University](https://huggingface.co/MCG-NJU/videomae-large) Released by Multimedia Computing Group-Nanjing University through its official repository on Hugging Face. [Read the license](https://creativecommons.org/licenses/by-nc/4.0/). ## Built From - Described by [arXiv:2111.06377](https://savrn.com/papers/masked-autoencoders-are-scalable-vision-learners) - Described by [arXiv:2203.12602](https://savrn.com/papers/videomae-masked-autoencoders-are-data-efficient-learners-for-self-supervised-video-pre-tra) ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 2.7 GB | | 16-bit | 0.7 GB | | 8-bit | 0.3 GB | | 4-bit | 0.2 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Built on This Model - Derived from[kandinsky-videomae-large-camera-motion](https://savrn.com/models/kandinsky-videomae-large-camera-motion) - Derived from[videomae-large-finetuned-deepfake-subset](https://savrn.com/models/videomae-large-finetuned-deepfake-subset) ## Questions About videomae-large ### How much GPU memory does videomae-large need? About 0.8 GB at 16-bit and 0.2 GB at 4-bit: the weights (343M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run videomae-large on? 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. ### Can I use videomae-large commercially? Not without separate permission. videomae-large is released under Creative Commons Attribution-NonCommercial 4.0. CC BY-NC 4.0 permits sharing and adapting with credit for non-commercial purposes only. Commercial use needs separate permission from the rights holder. ## Similar Models Model · Video classification ### [vjepa2.1-vit-large-384](https://savrn.com/models/vjepa2-1-vit-large-384) [Antonio Apicella](https://savrn.com/model-publishers/apiantonio) A HuggingFace-format conversion of Meta AI's V-JEPA 2.1 ViT-L/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. An equivalent community port already exists (Dev-Jahn/vjepa2.1-vitl-fpc64-384). This repository adds an independently reproduced conversion together with the validation results below. Forward outputs match the existing port to all reported digits. The only structural change is that the fused QKV projection of each attention block is split into separate query / key / value… Open weights mit 328M parameters transformers [View model](https://savrn.com/models/vjepa2-1-vit-large-384) Model · Video classification ### [vjepa2-vitl-fpc64-256](https://savrn.com/models/vjepa2-vitl-fpc64-256) [AI at Meta](https://savrn.com/model-publishers/facebook) 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… Open weights mit 326M parameters transformers [View model](https://savrn.com/models/vjepa2-vitl-fpc64-256) Model · Video classification ### [vjepa2-vitl-fpc32-256-diving48](https://savrn.com/models/vjepa2-vitl-fpc32-256-diving48) [AI at Meta](https://savrn.com/model-publishers/facebook) 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-L 256 model with video classification head pretrained on Diving 48 dataset. To run V-JEPA 2 model, ensure you have installed the latest transformers Open weights mit 375M parameters transformers [View model](https://savrn.com/models/vjepa2-vitl-fpc32-256-diving48) Model · Video classification ### [vjepa2-vitl-fpc16-256-ssv2](https://savrn.com/models/vjepa2-vitl-fpc16-256-ssv2) [AI at Meta](https://savrn.com/model-publishers/facebook) 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-L 256 model with video classification head pretrained on Something-Something-V2 dataset. To run V-JEPA 2 model, ensure you have installed the latest transformers Open weights mit 375M parameters transformers [View model](https://savrn.com/models/vjepa2-vitl-fpc16-256-ssv2) Model · Video classification ### [vjepa2-vitl-fpc16-256-ssv2](https://savrn.com/models/vjepa2-vitl-fpc16-256-ssv2-2) [Pavel Iakubovskii](https://savrn.com/model-publishers/qubvel-hf) 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 model with video classification head pretrained on Something-Something-V2 dataset. To run V-JEPA 2 model, ensure you have installed the latest transformers Open weights mit 375M parameters transformers [View model](https://savrn.com/models/vjepa2-vitl-fpc16-256-ssv2-2) Model · Video classification ### [videomae-large-finetuned-kinetics](https://savrn.com/models/videomae-large-finetuned-kinetics) [Multimedia Computing Group-Nanjing University](https://savrn.com/model-publishers/mcg-nju) VideoMAE model pre-trained for 1600 epochs in a self-supervised way and fine-tuned in a supervised way on Kinetics-400. It was introduced in the paper VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training by Tong et al. and first released in this repository. Disclaimer: The team releasing VideoMAE did not write a model card for this model so this model card has been written by the Hugging Face team. VideoMAE is an extension of Masked Autoencoders (MAE) to video. The architecture of the model is very similar to that of a standard Vision Transformer (ViT), with a decoder on top for predicting pixel values for masked patches. Videos are presented to… Open weights cc-by-nc-4.0 304M parameters transformers [View model](https://savrn.com/models/videomae-large-finetuned-kinetics) ## Multimedia Computing Group-Nanjing University [All models and datasets](https://savrn.com/model-publishers/mcg-nju) ## Versions - [12da269a02d3](https://savrn.com/models/videomae-large/versions/12da269a02d3) · current 2026-09-18 ## Explore More - [All video classification models](https://savrn.com/models/tasks/video-classification) - [All models under cc-by-nc-4.0](https://savrn.com/models/licenses/cc-by-nc-4-0) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-21. - [Hugging Face record](https://huggingface.co/MCG-NJU/videomae-large) - [How the hub is built](https://savrn.com/model-hub/methodology)