Modellix
今すぐ始める

MiniMax AIモデルシリーズ

MiniMax's video generation models are a family of multimodal video systems that turn text, images, and audio into high-quality, cinematic clips with native synchronized sound, strong motion, and up to 2K resolution.

8 モデル更新日 2026年10月

すべてのMiniMaxモデル

minimax/minimax-h3-max-v2v

minimax/minimax-h3-max-v2v

動画から動画

MiniMax's fast video-to-video model for drafts guided by reference motion, with optional stills and audio, and picture and sound together. It uses 1 to 3 reference videos in a 5 to 15 second clip at 480p or 768p.

minimax/minimax-h3-max-i2v

minimax/minimax-h3-max-i2v

画像から動画

MiniMax's fast reference-to-video model for drafts that keep a subject or style consistent, with picture and sound together. It uses 1 to 9 reference images and up to three audio clips, in a 5 to 15 second clip at 480p or 768p.

minimax/minimax-h3-max-fl2v

minimax/minimax-h3-max-fl2v

画像から動画

MiniMax's fast first-and-last-frame model for drafts that open on a still, land on an end frame, or bridge both, with picture and sound together. It makes a 5 to 15 second clip at 480p or 768p.

minimax/minimax-h3-max-t2v

minimax/minimax-h3-max-t2v

テキストから動画

MiniMax's fast text-to-video model for prompt-only drafts and short scenes, with picture and sound together. It makes a 5 to 15 second clip at 480p or 768p, including vertical 9:16 and ultrawide 21:9 frames.

minimax/minimax-h3-v2v

minimax/minimax-h3-v2v

動画から動画

MiniMax's video-to-video model for a new clip guided by reference motion, with optional stills and audio, and picture and sound together. It uses 1 to 3 reference videos in a 4 to 15 second clip at 768p or 2K.

minimax/minimax-h3-i2v

minimax/minimax-h3-i2v

画像から動画

MiniMax's reference-to-video model for a new scene that keeps characters, products, or a style consistent, with picture and sound together. It uses 1 to 9 reference images and up to three audio clips, in a 4 to 15 second clip at 768p or 2K.

minimax/minimax-h3-fl2v

minimax/minimax-h3-fl2v

画像から動画

MiniMax's first-and-last-frame model for opening on a still, landing on an end frame, or bridging both, with picture and sound together. It makes a 4 to 15 second clip at 768p or 2K.

minimax/minimax-h3-t2v

minimax/minimax-h3-t2v

テキストから動画

MiniMax's text-to-video model for storyboards, character shorts, and cinematic b-roll from a prompt, with picture and sound together. It makes a 4 to 15 second clip at 768p or 2K, including vertical 9:16 and ultrawide 21:9 frames.

必要なモデルが見つかりませんか? ご要望をお聞かせください。

さらに探す