Wan 2.7 is the newest model in Alibaba's video lineup — better temporal consistency, motion naturalness, and prompt adherence than Wan 2.6. The Wan family's most capable video model currently on ZenCreator.
🎵 Audio + video in one pass
Native audio generation alongside the video — ambient sound, music, and synchronized audio from a single prompt. No separate audio step required for most content.
⏱ Up to 15 seconds
5, 10, or 15-second clips — enough for short-form storytelling, product demos, and social content without cutting scenes mid-action. The full 15s range at 1080p in a single generation.
📺 1080p native
Full 1080p output — publish-ready for Reels, TikTok, and YouTube Shorts without upscaling. Drop to 720p when you want faster generation and lower credit cost.
🔓 Unrestricted for trusted users
No safety filters on the animation. Wan 2.7 runs with minimal content restrictions for trusted users — the same unrestricted access as the full Wan family on ZenCreator.
📌 Start frame anchoring
Upload any photo as the start frame and Wan 2.7 animates outward from it — preserving your character's face, outfit, and scene composition through the full clip.
What is WAN 2.7 Spicy?
Wan 2.7 is Alibaba Tongyi Lab's newest image-to-video model — the latest generation in a family that includes Wan 2.2, 2.5, and 2.6. Each version improved on temporal consistency and motion quality; Wan 2.7 raises the bar again with better prompt adherence, smoother inter-frame coherence, and more natural motion physics across a wider range of subject types.
Practically: upload a source image as the start frame, write a motion + audio prompt, and Wan 2.7 generates a 5–15 second clip at up to 1080p with native audio baked in. No end frame control (for start+end, use Wan 2.2). No LoRA style adaptation (for special styles, use Wan 2.2 + LoRAs).
The model is available for trusted users on ZenCreator with minimal content restrictions — the same unrestricted access as the full Wan family. Cost scales by resolution and duration: 720p is cheaper per second than 1080p.
Wan 2.7 is the most capable Wan video model on ZenCreator — the newest generation with native audio, 1080p output, and 15-second clips. Use it as the default choice when you need Alibaba's video quality plus audio in one generation. For start+end frame control, use Wan 2.2. For special LoRA styles, use Wan 2.2 + LoRAs.
Available in
Image-to-Video
Upload a source image, write a motion + audio prompt, pick Wan 2.7, generate at 720p or 1080p.
Wan 2.7 is a newer generation with improved temporal consistency, motion naturalness, and prompt adherence. Both support audio and 15-second duration at 1080p. Wan 2.7 is the more capable model; Wan 2.6 + Audio remains available as an alternative if you prefer it.
No — start frame only. You upload a photo as the opening frame and Wan 2.7 animates outward from it. For start AND end frame control (useful for loops and transitions), use Wan 2.2 or Kling 2.1.
Audio is generated automatically from your prompt — describe the sonic context (ambient rain, upbeat background track, crowd murmur) and the model matches it. Voice auto-selects in English. For a specific voice or non-English speech, generate silently and add audio via the Lipsync tool.
720p generates faster and costs fewer credits per second; 1080p produces sharper, publish-ready output. Use 720p for drafts and direction checks, switch to 1080p for the final clip you're posting.
Yes — Wan 2.7 runs with minimal content filters for trusted users on ZenCreator. No safety filters limit the animation output.
Both are premium unrestricted models with audio at 1080p. Wan 2.7 goes up to 15s and is the newer Alibaba model. Seedance Pro 1.5 (ByteDance) caps at 10s but adds explicit camera control (dolly, pan, orbit). Choose Wan 2.7 for longer clips; Seedance Pro 1.5 for precise camera movement.
