reference-to-video

AI Agent 驱动的开源可自部署视频工作台：将小说与剧本转为角色、场景、道具资产、分镜、视频和剪映草稿，支持跨镜头一致性、多供应商与费用追踪 | Self-hosted AI video workspace for stories, storyboards and short-form video production

[ECCV 2026] Official PyTorch implementation of RefAlign: Representation Alignment for Reference-to-Video Generation

Premiere-inspired PySide6 director studio for MiniMax H3 Ref2VA with AI shot planning, semantic media enrichment, timeline prompt reconciliation and shot-aware long-video rendering through ComfyUI.

Direct MiniMax H3 like a pro: Idea in, finished video out. A complete MiniMax H3 studio that lives in your ComfyUI.

MiniMax H3 全域提示词工程 Skill v1.8.1｜AIMixer Director 导演台生产规则 + 武打/兵器/枪斗编排｜T2VA / I2VA / FL2VA / L2VA / Ref2VA｜Subject 与 Picture 语义分离、跨镜头因果桥、人物一致性、动作参考、音频控制、Prompt 修复｜ComfyUI / RunningHub / Codex

MiniMax H3 video + Qwen Image 2.1. Native

Original MiniMax H3 prompt library, MuAPI Python client, API examples, and current model notes.

ComfyUI node pack for MiniMax H3 — chains generations into one continuous clip with held identity

ComfyUI custom nodes for Veo 3.1 video generation — text-to-video, image-to-video, reference-to-video, extend, and 4K upscale via MuAPI

MiniMax H3 (Hailuo 3.0) front-end for ComfyUI — text-to-video, image-to-video, first-and-last-frame and reference-to-video with native audio. A local LLM writes H3's structured prompts and self-corrects them. Character consistency from a subject photo, live preview, VRAM handoff.

ComfyUI custom nodes for MiniMax H3 text-to-video, image-to-video, and multimodal reference-to-video generation through Muapi.

Quality-first local MiniMax H3 video-series generation and loopback API for dual RTX 4090 workstations, with native audio, references, P8/P9 continuity, and preserved artifacts.

本地 MiniMax H3 参考生视频(R2V)竖屏玩龙工作流 —— ComfyUI，8GB 显存可跑，含 <Picture 1> 参考图绑定用法

MiniMax H3 本地视频生成工作台：T2V/I2V/R2V 三模式 + 多镜头时间线，单端口 FastAPI + Jinja2，一键启动 / Docker 部署 | Local MiniMax H3 video studio: T2V/I2V/R2V, multi-shot timeline, ComfyUI/diffusers engines, FastAPI + Jinja2

Pinned, hash-verified Comfy-Org MiniMax H3 ComfyUI workflows for T2V, I2V and R2V, with RTX 3060 benchmark evidence.

ComfyUI MiniMax H3 workflows: T2V, I2V, Ref2V, variable duration and LM Studio prompts / 音声付き動画・可変尺・日本語入力

50 MiniMax H3 prompts with the clips they produced - 40 curated from their creators, 10 rendered and measured here. Plus the prompt format MiniMax actually documents, and what the three upstream lanes really return.

A production-ready Seedance 2.5 prompt skill for text-to-video, image-to-video, reference-to-video, editing, extension, camera direction, and synchronized audio.

ComfyUI-RefCap: Reference capture for MiniMax H3. One file per character, object or place that replays H3's reference conditioning exactly.

Vidu — independent third-party profile of a public API surface, by API Evangelist. Vidu is a generative video AI platform from Shengshu Technology (ShengShu / 生数科技), built on the company's U-ViT diffusion-transformer architecture. The Vidu API turns text prompts, still images, and reference subjects into short video clips with features like text-to