{"paper":{"title":"VideoCrafter1: Open Diffusion Models for High-Quality Video Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Open diffusion models generate realistic videos at 1024x576 resolution from text, with an image-to-video version that preserves input content.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chao Weng, Haoxin Chen, Jinbo Xing, Menghan Xia, Qifeng Chen, Shaoshu Yang, Xiaodong Cun, Xintao Wang, Yaofang Liu, Yingqing He, Ying Shan, Yong Zhang","submitted_at":"2023-10-30T13:12:40Z","abstract_excerpt":"Video generation has increasingly gained interest in both academia and industry. Although commercial tools can generate plausible videos, there is a limited number of open-source models available for researchers and engineers. In this work, we introduce two diffusion models for high-quality video generation, namely text-to-video (T2V) and image-to-video (I2V) models. T2V models synthesize a video based on a given text input, while I2V models incorporate an additional image input. Our proposed T2V model can generate realistic and cinematic-quality videos with a resolution of $1024 \\times 576$, "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our proposed T2V model can generate realistic and cinematic-quality videos with a resolution of 1024 × 576, outperforming other open-source T2V models in terms of quality. The I2V model is the first open-source I2V foundation model capable of transforming a given image into a video clip while maintaining content preservation constraints.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that the models achieve the stated quality, outperformance, and strict content preservation, which rests on unspecified training details, evaluation metrics, and comparisons not provided in the abstract.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Open-source text-to-video and image-to-video diffusion models generate high-quality 1024x576 videos, with the I2V variant claimed as the first to strictly preserve reference image content.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Open diffusion models generate realistic videos at 1024x576 resolution from text, with an image-to-video version that preserves input content.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"26a43eb03499b864a0cf5d400d1688a7040ecd67c2e6b5b63740e5dd9c1e7cdf"},"source":{"id":"2310.19512","kind":"arxiv","version":1},"verdict":{"id":"233f9a2f-3823-47ad-929c-7ae7a25b849d","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T21:36:24.597386Z","strongest_claim":"Our proposed T2V model can generate realistic and cinematic-quality videos with a resolution of 1024 × 576, outperforming other open-source T2V models in terms of quality. The I2V model is the first open-source I2V foundation model capable of transforming a given image into a video clip while maintaining content preservation constraints.","one_line_summary":"Open-source text-to-video and image-to-video diffusion models generate high-quality 1024x576 videos, with the I2V variant claimed as the first to strictly preserve reference image content.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that the models achieve the stated quality, outperformance, and strict content preservation, which rests on unspecified training details, evaluation metrics, and comparisons not provided in the abstract.","pith_extraction_headline":"Open diffusion models generate realistic videos at 1024x576 resolution from text, with an image-to-video version that preserves input content."},"references":{"count":63,"sample":[{"doi":"","year":2023,"title":"Accessed October 22, 2023 [Online] https:// research.runwayml.com/gen2","work_id":"13424273-e098-4752-b508-3664d253c055","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Accessed October 22, 2023 [Online] https : / / github.com/deep-floyd/IF","work_id":"1f374479-4139-4a1a-9a9b-36e1914431f7","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Accessed October 22, 2023 [Online] https: //laion.ai/blog/laion-coco/","work_id":"77fb37fc-1c38-4984-ac70-290ea3a3b8c1","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Accessed October 22, 2023 [Online] https: //github.com/hotshotco/Hotshot-XL","work_id":"9a07b936-6609-4d38-8bd9-c271e4c8bba5","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Accessed October 22, 2023 [Online] https: //moonvalley.ai/","work_id":"23be4b42-07a0-452b-a925-35d30574fbd8","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":63,"snapshot_sha256":"4addebc6f30084e3d112e0c18f8750dba7b98813cb8eaecbd54a6dd2cbb31f4c","internal_anchors":14},"formal_canon":{"evidence_count":2,"snapshot_sha256":"48c27647a0639863ae022c6764e3894f10c0bbb17d182d8d29395c633332967b"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}