Introduction
Customized video generation combines multimodal reference inputs with the HunyuanVideo foundation.
View full image ↗Original method figure from the official HunyuanCustom project page, showing multimodal references and customised video generation.
Tencent HunyuanCustom research team · Rights in the original image and document remain with their respective holders
Source · HunyuanCustom: A Multimodal-Driven Architecture for Customized Video Generation ↗RESEARCH ACCOUNT
Subject consistency becomes an explicit goal of video customisation.
ANTI-AI ARCHIVE · Revised 2026-10-02
Release and use
HunyuanCustom conditions video generation around specified subjects. The parent repository records 9 May, while its own repository dates weights and inference code to 8 May; audio- and video-driven code follows in June. These dates should remain distinct rather than implying every research capability was available on the first day.
Section sources: HunyuanCustom: A Multimodal-Driven Architecture for Customized Video Generation
Change in creative practice
For continuous storytelling, a subject’s identity can matter more than single-frame polish. Customisation makes that requirement an input condition while raising the question of who may supply a person’s or object’s references. A consistency metric and a performer’s permission are different matters; neither substitutes for the other.
Section sources: HunyuanCustom: A Multimodal-Driven Architecture for Customized Video Generation
Sources for this account
Official release scope and demonstration descriptions read; vendor performance claims are not treated as independent test results.
This account expands the archive within the cited scope. Historical interpretation is not presented as the original author’s statement or as a proven causal influence. The account and translation are AI-assisted, pending independent human review. Section references identify evidence without claiming independent verification of every historical statement.
Continue with a comparative question
Compare the methods, forms of control or presentation contexts of this entry and HunyuanVideo.
Compare the methods, forms of control or presentation contexts of this entry and HunyuanVideo-Avatar.
Dates & version record
Date displayed for this node: 2025-05-09 · Historical date recorded for the source: 2025-05-09
These dates refer to the historical event or recorded version, not this page’s publication date. The original date precision and unresolved questions are retained.
Original sources & further reading
These links lead to the cited paper, article, institution or conference page. External texts retain their source languages.
This is the archive’s bilingual research abstract. The external link is an original source, not a claim that the publisher offers full text in both languages.
Provenance, translation & verification
Timeline entry #291 · Open model
Research materials & supplement references · 1
GenAI Visual Release Chronology Research Master · p. 30 · R163
Date as recorded: 2025-05-09
HunyuanCustom ↗
HunyuanCustom
Archive account based on the listed research materials, not a full translation of the linked work. AI-assisted translation; independent human review pending.
The master’s research claims and dates await independent verification.
Cite this node
ANTI-AI ARCHIVE. HunyuanCustom. Art-history node #291. https://salondesrefuses.cn/en/art-history/200
For specific historical claims, also cite the original sources above and include your access date. This account is not a full translation of the linked work.
Adjacent nodes follow chronological order; adjacency does not establish direct influence or causation.