現在、世界中のすべての公開コミュニティメンバーとクリエイティブクリエイターが利用可能ですMarch 2026

Wan 2.7: Advanced AI Video Generator with Precision Controls

Alibaba’s updated Wan 2.7 AI video model delivers industry-leading first/last frame control, multi-reference input support, and intuitive natural language editing. Create polished 5 to 15-second video clips at 720P or 1080P resolution.

Omni Video ProのDiscordコミュニティに参加する

Omni Video Proコンテンツポリシーの通知

入力が規制に違反すると生成に失敗します。実写の人物、NSFWコンテンツ、暴力的なコンテンツ、著作権で保護されたコンテンツは、モデルのセーフティフィルターでブロックされる場合があります。スタイライズドアート、架空のキャラクター、AI生成の被写体は通常、Omni Video Proで正常に生成されます。

Key Wan 2.7 Standout Capabilities

Why Wan 2.7 Stands Out as a Top AI Video Generator

Wan 2.7 elevates Alibaba’s video generation toolkit with first/last frame locking, multi-reference input support, natural language editing tools, and support for clips up to 15 seconds in length.

Precise First & Last Frame Locking

Lock in your perfect opening and closing visual frames before starting generation. Wan 2.7 automatically builds smooth, polished motion between those two fixed points, putting precise cinematic creative control at your fingertips without needing complex prompt workflows.

Lock in both your starting and ending visual compositions before starting the generation process.

Perfect for product reveals, character transitions, and crisp, clean scene shifts.

Eliminates the guesswork of hitting a precise final visual target.

Multi-Reference Video Support

Upload up to 5 reference videos at once to guide the model’s character design, scene environment, and overall motion style.

Upload up to 5 reference clips to perfectly tailor your final video output.

Keep consistent visual styling for characters and environments throughout your full clip.

Ideal for brand marketing campaigns, fashion reels, and maintaining product consistency in commercial video projects.

Natural Language-Powered Video Editing

Edit existing video clips using straightforward, plain natural language prompts. Swap backgrounds, adjust lighting, modify clothing, or refine your video’s style without rebuilding the entire clip from scratch.

List your desired changes using simple text — no advanced timeline editing skills needed.

Swap backgrounds, update character outfits, or adjust lighting with just one prompt.

Quickly iterate on your footage without losing the original clip’s core motion and timing.

Extended 15-Second Clip Durations

Create clips up to 15 seconds in length — three times longer than prior Wan video models. This increased duration is perfect for full product demos or short, standalone cinematic sequences.

Choose clip lengths of 5, 10, or 15 seconds to match your project’s unique needs.

Offers 480P, 720P, and 1080P output resolutions for a wide range of use cases.

Works with both 16:9 landscape and 9:16 portrait aspect ratios for complete flexibility.

Discover Additional AI Video Tools

よくある質問

Omni Video Pro、Google Omni AI Video、および現在の生成AIビデオ生成サポートについて

What is Wan 2.7?

Wan 2.7 is Alibaba Tongyi Lab’s top-tier video generation model, initially launched in March 2026. Building directly on the Wan 2.6 framework, this updated release includes game-changing upgrades: first/last frame locking, support for up to 5 concurrent reference videos, 9-grid image inputs, natural instruction-based editing, and refined motion physics for smoother, more lifelike final outputs.

What is first/last frame control in Wan 2.7?

First/last frame control (often shortened to FLF2V) lets you lock in both the opening and closing visual frames for your generated video. Wan 2.7 automatically generates seamless, polished motion between those two fixed points, putting full cinematic creative control directly at your disposal. Lock in your perfect opening and ending compositions, then let the model handle the transitional middle footage.

How long can videos be with Wan 2.7?

Each clip generated with Wan 2.7 spans 2 to 15 seconds total — a major leap from earlier Wan models that maxed out at roughly 5 seconds. On this platform, you can choose 5, 10, or 15-second clip lengths to fit your project’s specific needs.

What modes does Wan 2.7 support?

Wan 2.7 supports image-to-video, text-to-video, first/last frame video (FLF2V), and natural language-driven video editing. Currently, this platform’s live implementation grants you full access to image-to-video and text-to-video as its active, fully functional modes.

What resolutions does Wan 2.7 support?

Wan 2.7 can export video at 480P, 720P, and 1080P resolutions. Both 16:9 landscape and 9:16 portrait aspect ratio settings are fully supported, letting you adapt to any project’s format needs.

Is Wan 2.7 open source?

The prior Wan 2.1 model was fully open-sourced under the Apache-2.0 license. When Wan 2.7 first launched, official open-source specifications had not been finalized. For the latest, most up-to-date details, visit the Alibaba Wan GitHub repository at github.com/Wan-Video.

How does Wan 2.7 compare to Wan 2.6?

Standout upgrades from Wan 2.6 to Wan 2.7 include first/last frame locking, 9-grid multi-image input support, up to 5 concurrent reference video inputs, and natural instruction-based editing — all capabilities missing from the previous Wan 2.6 model. Clip duration was also extended to a 15-second maximum, and both motion physics accuracy and character visual consistency saw substantial, noticeable enhancements.

Omni Video Proについてまだ質問がありますか？専任のプロサポートチームがあなたをサポートする準備ができています。