Grok Imagine Video 1.5 Promises Better Motion and Built-In Audio, Twice as Fast

xAI shipped Grok Imagine Video 1.5 yesterday, and the pitch is the one every AI video tool eventually has to make: better quality, and faster than before.

The model is now generally available on the Imagine API, with a faster version, Video 1.5 Fast, rolling out on grok.com/imagine and xAI's iOS and Android apps. xAI calls these its best image-to-video models yet, citing improvements to motion, physics, audio, and generation speed. xx

What's actually new

Sound effects, ambience, and dialogue now generate in the same pass as the video itself, landing on the action rather than being layered on afterward, and speech is described as clearer and better synced. That's a meaningful workflow change if accurate. Most AI video tools still treat audio as a separate generation step, or skip it and leave creators to add their own.

xAI also says movement now holds together better over the length of a clip, with fewer warps and more believable weight and momentum, a problem that has dogged nearly every AI video model to date, where limbs and objects tend to drift or deform a few seconds into a clip.

On speed, the company says Grok Imagine Video 1.5 Fast nearly doubles generation speed compared to the previous model, producing a 6-second, 720p video in about 25 seconds, down from 40-plus seconds before.

New workflow features

Alongside the model update, xAI is rolling out three workflow features over the next few days. Projects let users organize generations into folders that appear in the sidebar. Multiple agents can now run in parallel on a project, letting users kick off several prompts at once instead of waiting for each generation to finish before starting the next. A new search feature lets users find any previously generated image or video in their library without scrolling.

The parallel-agent feature in particular addresses one of the more tedious parts of AI video generation: the wait-and-iterate cycle. Letting several prompt variations run simultaneously instead of sequentially could meaningfully speed up the process of finding a usable take.

How developers can use it

Grok Imagine Video 1.5 is out of preview and generally available in the xAI API under the model name grok-imagine-video-1.5. Developers feed it a starting image, describe the motion they want, and choose resolution and duration.

Competitive Context

xAI is entering a field that's gotten considerably more crowded over the past year, with Google's Veo, OpenAI's Sora, and Runway's Gen-series models all competing on similar ground: image-to-video generation with increasingly convincing motion and physics.

The specific claim worth watching here is the built-in audio generation, since most rival tools either generate video and audio as separate passes or don't generate audio at all, leaving sound design to the creator afterward. If xAI's same-pass approach holds up to real testing, it's a meaningful differentiator rather than just a speed claim.

The Signal in the Noise

The speed claim is the easiest one to verify and the one creators will notice immediately. Cutting generation time on a short clip from 40-plus seconds to roughly 25 is a real, usable improvement for anyone iterating on multiple takes, regardless of how the other quality claims hold up under scrutiny.

The harder claims to evaluate from a press release alone are the motion and physics improvements, since "fewer warps" and "more believable weight" are exactly the kind of claims that need hands-on testing against specific prompts before they mean much. Early hands-on comparisons from outlets that test multiple prompts side by side against the previous version, and against Veo and Sora, will tell the real story here.

The parallel-agents feature is the most interesting structural change, less because of the underlying model and more because it nudges the entire workflow toward treating generation as a batch process rather than a one-at-a-time loop. That's a meaningful shift in how creators might actually use these tools day to day.

Specs & Pricing

xAI has not published a single, unified per-second or per-minute price specifically for Grok Imagine Video 1.5 on its own pricing page as of this writing. Exact current pricing for both the consumer app and the API is available on xAI's website.

Resources & Reads

xAI's full technical documentation for the Imagine API, including endpoints, parameters, and authentication, covers the complete developer integration path for Grok Imagine Video 1.5.