Text to video AI: models and plans

Text to video in Quad Studio

Quad Chat has 17 text-to-video models from ByteDance, Google, Kling AI, Alibaba, Black Forest Labs, xAI and 1 more in Quad Studio. Describe a scene, choose a model, length, resolution and aspect ratio, and the clip lands in your library. Studio shows each model's supported lengths, resolutions and aspect ratios before you generate, and an estimate of the credits the clip will use. Video starts on the Ultra plan ($40/month) with Seedance 1 Pro Fast, Veo 3.1 Lite, Kling 2.6, Wan 3 Video, Grok Imagine Video 1.5 Lite and Seedance 2.0 Mini. Max ($200/month) adds Seedance 1.5 Pro, Wan 3 Video Prime, FLUX 3 Video, Grok Imagine Video, Grok Imagine Video 1.5, Veo 3.1 Fast and 5 more. The Free, Go and Pro plans do not include video.

Text to video models compared

Clip length is the shortest to the longest the model supports. Click a header to sort. Seedance 1 Pro Fast (Ultra, 2 to 12 seconds, up to 1080p); Seedance 1.5 Pro (Max, 4 to 12 seconds, up to 1080p, audio); Veo 3.1 Lite (Ultra, 4 to 8 seconds, up to 1080p, audio); Kling 2.6 (Ultra, 5 to 10 seconds, up to 720p); Wan 3 Video (Ultra, 2 to 30 seconds, up to 1080p); Wan 3 Video Prime (Max, 2 to 30 seconds, up to 1080p); FLUX 3 Video (Max, 5 to 20 seconds, up to 720p, audio); Grok Imagine Video (Max, 1 to 15 seconds, up to 720p); Grok Imagine Video 1.5 (Max, 1 to 15 seconds, up to 1080p, audio); Veo 3.1 Fast (Max, 4 to 8 seconds, up to 1080p, audio); MiniMax H3 (Max, 4 to 15 seconds, up to 1440p, audio); MiniMax H3 Max (Max, 5 to 15 seconds, up to 768p); Kling 3 (Max, 3 to 15 seconds, up to 720p, audio); Veo 3.1 (Max, 4 to 8 seconds, up to 1080p, audio); Grok Imagine Video 1.5 Lite (Ultra, 1 to 15 seconds, up to 1080p, audio); Seedance 2.0 Mini (Ultra, 4 to 15 seconds, up to 720p, audio); Seedance 2.5 (Max, 4 to 30 seconds, up to 1080p, audio).

How to choose a video model

Straight from the catalog, not from quality benchmarks. Longest clips: Wan 3 Video, up to 30 seconds. The shortest supported clip across models is 1 second. Highest resolution: MiniMax H3, up to 1440p. Fast generation: Seedance 1 Pro Fast, Veo 3.1 Lite, Kling 2.6, Wan 3 Video Prime and MiniMax H3 Max. Native audio: Seedance 1.5 Pro, Veo 3.1 Lite, FLUX 3 Video, Grok Imagine Video 1.5, Veo 3.1 Fast and 6 more. Turn sound on before you generate; clips with audio use more credits than silent ones. Image input: 14 of these 17 models can also animate a starting image; see the image-to-video page. Entry plan: Seedance 1 Pro Fast, Veo 3.1 Lite, Kling 2.6, Wan 3 Video and 2 more on Ultra ($40/month).

Credits, plans and privacy

Video shares one monthly credit balance with chat and images: Ultra 5,000 credits, Max 25,000. Longer, sharper clips cost more, and the estimate updates before you generate. Image and video models run on the Standard Lane, so the provider may retain prompts, reference images and results under its own data policy. Quad asks you to confirm this once before your first generation, and chat keeps the Private Lane on supported models.

Frequently asked questions

Which Quad Chat plan includes text-to-video generation?

Ultra ($40/month) is the first plan with Seedance 1 Pro Fast, Veo 3.1 Lite, Kling 2.6, Wan 3 Video and 2 more. Higher plans add Seedance 1.5 Pro, Wan 3 Video Prime, FLUX 3 Video, Grok Imagine Video, Grok Imagine Video 1.5 and 6 more. The Free, Go and Pro plans do not include video.

Which text-to-video models are available?

17 models from ByteDance, Google, Kling AI, Alibaba, Black Forest Labs, xAI and MiniMax, including Seedance 1 Pro Fast, Seedance 1.5 Pro, Veo 3.1 Lite, Kling 2.6, Wan 3 Video, Wan 3 Video Prime, FLUX 3 Video, Grok Imagine Video and 9 more. They sit next to chat and image models in one app.

How long can the clips be?

Length depends on the model, from 1 to 30 seconds (Wan 3 Video is the longest), at up to 1440p on MiniMax H3. Studio shows each model's supported lengths, resolutions and aspect ratios before you generate.

Can I make AI videos with sound?

Yes, on models with native audio: Seedance 1.5 Pro, Veo 3.1 Lite, FLUX 3 Video, Grok Imagine Video 1.5, Veo 3.1 Fast and 6 more. Turn sound on before you generate; clips with audio use more credits than silent ones, and the estimate updates first.

Can I start from an image instead of a prompt?

Yes. 14 video models accept a starting image, including Seedance 1 Pro Fast, Seedance 1.5 Pro, Wan 3 Video, Wan 3 Video Prime and 10 more. See the image-to-video page for the list.

Is my data private in Quad Studio?

Image and video models run on the Standard Lane, so the provider may retain prompts, reference images and results under its own data policy. Quad asks you to confirm this once before your first generation, and chat keeps the Private Lane on supported models.