{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/magvit-masked-generative-video-transformer","title":"MAGVIT: Masked Generative Video Transformer","arxiv_id":"2212.05199","date":"2022-12-10","proceeding":"CVPR 2023 1","authors":["Lijun Yu","Yong Cheng","Kihyuk Sohn","José Lezama","Han Zhang","Huiwen Chang","Alexander G. Hauptmann","Ming-Hsuan Yang","Yuan Hao","Irfan Essa","Lu Jiang"],"abstract":"We introduce the MAsked Generative VIdeo Transformer, MAGVIT, to tackle various video synthesis tasks with a single model. We introduce a 3D tokenizer to quantize a video into spatial-temporal visual tokens and propose an embedding method for masked video token modeling to facilitate multi-task learning. We conduct extensive experiments to demonstrate the quality, efficiency, and flexibility of MAGVIT. Our experiments show that (i) MAGVIT performs favorably against state-of-the-art approaches and establishes the best-published FVD on three video generation benchmarks, including the challenging Kinetics-600. (ii) MAGVIT outperforms existing methods in inference time by two orders of magnitude against diffusion models and by 60x against autoregressive models. (iii) A single MAGVIT model supports ten diverse generation tasks and generalizes across videos from different visual domains. The source code and trained models will be released to the public at https://magvit.cs.cmu.edu.","url_abs":"https://arxiv.org/abs/2212.05199v2","url_pdf":"https://arxiv.org/pdf/2212.05199v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"magvit-masked-generative-video-transformer","repo_url":"https://github.com/google-research/magvit","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"jax","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"text-to-video-generation","task_name":"Text-to-Video Generation"},{"task_slug":"video-generation","task_name":"Video Generation"},{"task_slug":"video-prediction","task_name":"Video Prediction"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"diffusion","method_name":"Diffusion"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/text-to-video-generation-on-something","task":"Text-to-Video Generation","dataset":"Something-Something V2","model":"MAGVIT","rank_in_archive_order":1,"of":1,"metrics":{"FVD":"79.1"},"uses_additional_data":false},{"leaderboard":"/sota/video-generation-on-bair-robot-pushing","task":"Video Generation","dataset":"BAIR Robot Pushing","model":"MAGVIT","rank_in_archive_order":1,"of":31,"metrics":{"Cond":"1","FVD score":"62","Pred":"15","Train":"15"},"uses_additional_data":false},{"leaderboard":"/sota/video-generation-on-kinetics-600-12-frames","task":"Video Generation","dataset":"Kinetics-600 12 frames, 64x64","model":"MAGVIT","rank_in_archive_order":3,"of":4,"metrics":{"FVD":"9.9"},"uses_additional_data":false},{"leaderboard":"/sota/video-generation-on-ucf-101","task":"Video Generation","dataset":"UCF-101","model":"MAGVIT (-L-CG, 128x128, class-conditional)","rank_in_archive_order":7,"of":48,"metrics":{"FVD16":"76±2","Inception Score":"89.27±0.15"},"uses_additional_data":false},{"leaderboard":"/sota/video-generation-on-ucf-101","task":"Video Generation","dataset":"UCF-101","model":"MAGVIT (-B-CG, 128x128, class-conditional)","rank_in_archive_order":12,"of":48,"metrics":{"FVD16":"159±2","Inception Score":"83.55±0.14"},"uses_additional_data":false},{"leaderboard":"/sota/video-generation-on-ucf-101","task":"Video Generation","dataset":"UCF-101","model":"MAGVIT (AR)","rank_in_archive_order":19,"of":48,"metrics":{"FVD16":"265"},"uses_additional_data":false},{"leaderboard":"/sota/video-prediction-on-bair-robot-pushing-1","task":"Video Prediction","dataset":"BAIR Robot Pushing","model":"MAGVIT (-L-FP)","rank_in_archive_order":1,"of":6,"metrics":{"FVD":"62±0.1"},"uses_additional_data":false},{"leaderboard":"/sota/video-prediction-on-bair-robot-pushing-1","task":"Video Prediction","dataset":"BAIR Robot Pushing","model":"MAGVIT (-B-FP)","rank_in_archive_order":2,"of":6,"metrics":{"FVD":"76±0.1"},"uses_additional_data":false},{"leaderboard":"/sota/video-prediction-on-kinetics-600-12-frames","task":"Video Prediction","dataset":"Kinetics-600 12 frames, 64x64","model":"MAGVIT (-L-FP)","rank_in_archive_order":5,"of":16,"metrics":{"Cond":"5","FVD":"9.9±0.3","Pred":"11"},"uses_additional_data":false},{"leaderboard":"/sota/video-prediction-on-kinetics-600-12-frames","task":"Video Prediction","dataset":"Kinetics-600 12 frames, 64x64","model":"MAGVIT (-B-FP)","rank_in_archive_order":9,"of":16,"metrics":{"Cond":"5","FVD":"24.5±0.9","Pred":"11"},"uses_additional_data":false},{"leaderboard":"/sota/video-prediction-on-something-something-v2","task":"Video Prediction","dataset":"Something-Something V2","model":"MAGVIT","rank_in_archive_order":1,"of":1,"metrics":{"FVD":"28.5"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2212.05199","atlas_url":"https://app.syntology.ai/?focus=2212.05199","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2212.05199"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/google-research/magvit","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":8,"ran_draft_wrong":1,"unverified":1},"by_repo_kind":{"official":{"samples":10,"ran":9,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"1a792a8e5a6e134c","entry":"frame_interpolation_cond","repo":"google-research/magvit","repo_kind":"official","path":"videogvt/train_lib/cond_utils.py","file_url":"https://github.com/google-research/magvit/blob/HEAD/videogvt/train_lib/cond_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1a792a8e5a6e134c"}},{"code_sha256_prefix":"0f46f383ab81bf4b","entry":"frame_prediction_cond","repo":"google-research/magvit","repo_kind":"official","path":"videogvt/train_lib/cond_utils.py","file_url":"https://github.com/google-research/magvit/blob/HEAD/videogvt/train_lib/cond_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0f46f383ab81bf4b"}},{"code_sha256_prefix":"b87899e786e53909","entry":"get_norm_layer","repo":"google-research/magvit","repo_kind":"official","path":"videogvt/models/model_utils.py","file_url":"https://github.com/google-research/magvit/blob/HEAD/videogvt/models/model_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b87899e786e53909"}},{"code_sha256_prefix":"6355c1dcf1ced438","entry":"l2_normalize","repo":"google-research/magvit","repo_kind":"official","path":"videogvt/models/vqvae.py","file_url":"https://github.com/google-research/magvit/blob/HEAD/videogvt/models/vqvae.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"6355c1dcf1ced438"}},{"code_sha256_prefix":"d73c517c7648170b","entry":"latent_frame_prediction_cond","repo":"google-research/magvit","repo_kind":"official","path":"videogvt/train_lib/cond_utils.py","file_url":"https://github.com/google-research/magvit/blob/HEAD/videogvt/train_lib/cond_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d73c517c7648170b"}},{"code_sha256_prefix":"0fd41694a130f120","entry":"sinusoidal_init","repo":"google-research/magvit","repo_kind":"official","path":"videogvt/models/transformer_lm.py","file_url":"https://github.com/google-research/magvit/blob/HEAD/videogvt/models/transformer_lm.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0fd41694a130f120"}},{"code_sha256_prefix":"4d2dd72078f90cca","entry":"truncated_normal","repo":"google-research/magvit","repo_kind":"official","path":"videogvt/models/simplified_bert.py","file_url":"https://github.com/google-research/magvit/blob/HEAD/videogvt/models/simplified_bert.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4d2dd72078f90cca"}},{"code_sha256_prefix":"0c839479ca2b7173","entry":"truncated_normal","repo":"google-research/magvit","repo_kind":"official","path":"videogvt/models/t5_bert.py","file_url":"https://github.com/google-research/magvit/blob/HEAD/videogvt/models/t5_bert.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0c839479ca2b7173"}},{"code_sha256_prefix":"ac4f595178d50862","entry":"vmap_t_dim","repo":"google-research/magvit","repo_kind":"official","path":"videogvt/models/model_utils.py","file_url":"https://github.com/google-research/magvit/blob/HEAD/videogvt/models/model_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ac4f595178d50862"}},{"code_sha256_prefix":"38e696a22dc81a6f","entry":"tensorflow_style_avg_pooling","repo":"google-research/magvit","repo_kind":"official","path":"videogvt/models/model_utils.py","file_url":"https://github.com/google-research/magvit/blob/HEAD/videogvt/models/model_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"38e696a22dc81a6f"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}