Wan 2.7 - Text to Video

Generate video from a text prompt with Wan 2.7, Alibaba's latest video generation model.

Write a prompt describing your scene, pick a resolution and duration, and run it. The workflow handles everything else. Your video comes back ready to preview and download.

How do you generate video with Wan 2.7?

Write a scene description in the prompt field, choose your resolution and aspect ratio, set your duration, and click Run. Audio input is optional if you want sound-synced generation.

Prompt Describe your scene with specific visual details. Shot framing, lighting, motion, and style all help the model understand what you want. Wan 2.7 responds well to structured prompts that break the scene into beats or shots. Keep your descriptions concrete rather than abstract.

Negative Prompt Tell the model what to avoid. "Low resolution, error, worst quality, low quality, deformed, extra fingers, bad proportions" is a solid starting point. Add anything specific to your scene that you want to steer away from.

Resolution Default is 1080P. Generate at 720P for faster drafting.

Ratio Set your aspect ratio. 16:9 is the default and works for most video content. Other options are available depending on your use case.

Duration Controls video length in seconds. Default is 10 seconds, but you can generate all the way up to 15 seconds.

Prompt Extend Enabled by default. This lets the model expand your prompt internally for better results. If you want exact control over what the model sees, turn it off.

Watermark Off by default. Toggle on if you want the Wan watermark on your output.

Seed Set to randomize by default. Lock a specific seed if you want to reproduce a result or compare prompt changes without other variables shifting.

What is Wan 2.7 text to video good for?

Wan 2.7 is Alibaba's newest video model, built for high quality text-to-video generation with audio support. It handles detailed scene descriptions well and produces coherent motion across multi-second clips at up to 1080P resolution.

This workflow is the fastest way to test Wan 2.7 on Floyo. No extra nodes, no complex setup. Write a prompt and generate.

It's a good fit for quick concept videos, scene previews, storyboard-to-motion tests, and exploring what the model can do with different prompt styles. If you need image-to-video or more control over the generation, look for dedicated Wan 2.7 I2V workflows instead.

The audio input is worth experimenting with. You can feed in audio and the model will attempt to sync the visual generation to it.

FAQ

What does Wan 2.7 cost to run on Floyo? Wan 2.7 runs through a Floyo API node, so the cost comes from your API Wallet rather than FloTime. The node displays the cost directly on the canvas before you run it. Longer durations and higher resolutions cost more.

How long does Wan 2.7 take to generate? Generation time depends on your duration and resolution settings. A 10-second clip at 1080P 16:9 takes roughly 5 to 7 minutes.

Can I use audio with Wan 2.7 text to video? Yes. There's an optional audio input on the API node. You can connect an audio file or provide a URL, and Wan 2.7 will factor the audio into the generation. Leave it empty if you don't need it.

What's the difference between Wan 2.7 and Wan 2.6? Wan 2.7 is Alibaba's latest release with improvements to generation quality and audio-visual capabilities. If you've been using earlier Wan versions, this is the upgrade path.