{"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/2505-11497","title":"QVGen: Pushing the Limit of Quantized Video Generative Models","arxiv_id":"2505.11497","date":"2025-05-16","proceeding":null,"authors":["Yushi Huang","Ruihao Gong","Jing Liu","Yifu Ding","Chengtao Lv","Haotong Qin","Jun Zhang"],"abstract":"Video diffusion models (DMs) have enabled high-quality video synthesis. Yet, their substantial computational and memory demands pose serious challenges to real-world deployment, even on high-end GPUs. As a commonly adopted solution, quantization has proven notable success in reducing cost for image DMs, while its direct application to video DMs remains ineffective. In this paper, we present QVGen, a novel quantization-aware training (QAT) framework tailored for high-performance and inference-efficient video DMs under extremely low-bit quantization (e.g., 4-bit or below). We begin with a theoretical analysis demonstrating that reducing the gradient norm is essential to facilitate convergence for QAT. To this end, we introduce auxiliary modules ($\\Phi$) to mitigate large quantization errors, leading to significantly enhanced convergence. To eliminate the inference overhead of $\\Phi$, we propose a rank-decay strategy that progressively eliminates $\\Phi$. Specifically, we repeatedly employ singular value decomposition (SVD) and a proposed rank-based regularization $\\mathbf{\\gamma}$ to identify and decay low-contributing components. This strategy retains performance while zeroing out inference overhead. Extensive experiments across $4$ state-of-the-art (SOTA) video DMs, with parameter sizes ranging from $1.3$B $\\sim14$B, show that QVGen is the first to reach full-precision comparable quality under 4-bit settings. Moreover, it significantly outperforms existing methods. For instance, our 3-bit CogVideoX-2B achieves improvements of $+25.28$ in Dynamic Degree and $+8.43$ in Scene Consistency on VBench.","url_abs":"https://arxiv.org/abs/2505.11497v1","url_pdf":"https://arxiv.org/pdf/2505.11497v1.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":[],"tasks":[{"task_slug":"quantization","task_name":"Quantization"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2505.11497","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2505.11497"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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":"deterministic:regex_extraction","url":"https://github.com/ModelTC/QVGen","reach":null}],"summary":{"unverified":7},"by_repo_kind":{"found_in_text":{"samples":7,"ran":0,"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":"7c3197c1e21d21a3","entry":"AdaSTE","repo":"ModelTC/QVGen","repo_kind":"found_in_text","path":"training/quant/quant.py","file_url":"https://github.com/ModelTC/QVGen/blob/HEAD/training/quant/quant.py","link_basis":"first_harvest_node","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":"7c3197c1e21d21a3"}},{"code_sha256_prefix":"e6172c8b0b846706","entry":"LSQPlusQuantizer","repo":"ModelTC/QVGen","repo_kind":"found_in_text","path":"training/quant/quant.py","file_url":"https://github.com/ModelTC/QVGen/blob/HEAD/training/quant/quant.py","link_basis":"first_harvest_node","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":"e6172c8b0b846706"}},{"code_sha256_prefix":"44c67898ae6ee3c9","entry":"LSQQuantizer","repo":"ModelTC/QVGen","repo_kind":"found_in_text","path":"training/quant/quant.py","file_url":"https://github.com/ModelTC/QVGen/blob/HEAD/training/quant/quant.py","link_basis":"first_harvest_node","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":"44c67898ae6ee3c9"}},{"code_sha256_prefix":"b8cdfdacd5e3b96a","entry":"LearnableAdaSTE","repo":"ModelTC/QVGen","repo_kind":"found_in_text","path":"training/quant/quant.py","file_url":"https://github.com/ModelTC/QVGen/blob/HEAD/training/quant/quant.py","link_basis":"first_harvest_node","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":"b8cdfdacd5e3b96a"}},{"code_sha256_prefix":"85bc07f29925aa0b","entry":"ProgressiveLoRAQuantizer","repo":"ModelTC/QVGen","repo_kind":"found_in_text","path":"training/quant/quant.py","file_url":"https://github.com/ModelTC/QVGen/blob/HEAD/training/quant/quant.py","link_basis":"first_harvest_node","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":"85bc07f29925aa0b"}},{"code_sha256_prefix":"a497cc143beaaa7c","entry":"UniformQuantizer","repo":"ModelTC/QVGen","repo_kind":"found_in_text","path":"training/quant/quant.py","file_url":"https://github.com/ModelTC/QVGen/blob/HEAD/training/quant/quant.py","link_basis":"first_harvest_node","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":"a497cc143beaaa7c"}},{"code_sha256_prefix":"3dba0daf72600cd7","entry":"low_rank_decomposition","repo":"ModelTC/QVGen","repo_kind":"found_in_text","path":"training/quant/quant.py","file_url":"https://github.com/ModelTC/QVGen/blob/HEAD/training/quant/quant.py","link_basis":"first_harvest_node","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":"3dba0daf72600cd7"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}