xAI Rolls Out Grok Imagine Video 1.5 Broadly, Pushing Faster Image-to-Video Generation
xAI has pushed its newest image-to-video model, Grok Imagine Video 1.5, out of preview and into broad availability this week, putting faster, audio-synced clip generation in front of developers, web users, and millions of people on the Grok mobile apps. The release, dated June 16 on xAI's site and amplified by the company on June 17, leans hard on a single pitch: better quality at meaningfully faster speeds, with the headline number being six-second 720p videos generated in about 25 seconds.
The move lands in the middle of an increasingly crowded AI video race, and it arrives alongside a quieter change to how much Grok Voice Mode subscribers actually get for their money.
What shipped
Grok Imagine Video 1.5 is now generally available through xAI's Imagine API, where it had previously sat in preview. Developers can call it as `grok-imagine-video-1.5`, hand it a starting image, describe the motion they want, and choose a resolution and duration. xAI simultaneously rolled out a lighter variant, Grok Imagine Video 1.5 Fast, on grok.com/imagine and the company's iOS and Android apps, so the update is not confined to the developer audience.
xAI describes the 1.5 line as "our best image-to-video models yet: better motion, better physics, better audio, at the fastest speeds." The company says the model improves "across every dimension that matters for real creative work" compared with its predecessor.
The most concrete gain is speed. According to xAI, Video 1.5 Fast "almost doubles generation speed," producing six-second 720p clips in roughly 25 seconds, down from 40-plus seconds in the prior model. That tighter loop matters more than it sounds: faster renders let creators throw away bad takes sooner instead of waiting through a long generation only to discover the timing, physics, or audio missed.
On quality, xAI points to two areas. Audio is now produced in the same pass as the video rather than bolted on afterward — sound effects, ambience, and dialogue are generated together and, per the company, "land on the action," with speech that is "clearer and better synced." On motion, xAI says movement "holds together over the length of a clip — fewer warps, more believable weight and momentum," a nod to the warped hands and objects that have long broken the illusion in generative video.
Alongside the model, xAI is rolling out creator-workflow features over the following days: Projects to organize work in a sidebar, the ability to run multiple agents in parallel so several prompts generate at once, and a library search so users can find a past clip without endless scrolling.
The Voice Mode change
Less prominent in xAI's messaging, but notable for paying users, is an adjustment to Grok Voice Mode allowances. Standard SuperGrok subscribers now get roughly 60 minutes of voice conversation per day, while the top-tier SuperGrok Heavy plan carries a 240-minute daily allowance. Reporting on the limits over the spring suggests the standard cap was trimmed from a previously higher allotment, part of a broader round of throttling that xAI applied to video, image, and voice features in May. Voice usage is metered in minutes of real-time conversation, with each minute reportedly consuming several message-equivalents — so a heavy talker can burn through a daily cap faster than the headline number implies.
Where it sits in the race
The video-generation field xAI is competing in has stratified into a handful of serious contenders. Trade coverage places Grok Imagine Video 1.5 near the top of the public Image-to-Video Arena leaderboard, with an Elo score in the neighborhood of 1,330 — meaning it is genuinely competitive on output quality, not just on price. The rivals it lines up against in that arena include ByteDance's Seedance, Kuaishou's Kling, and MiniMax's Hailuo, with Google's Veo and Runway's models anchoring the higher-production end of the market.
xAI's clearest lever against that group is cost and speed rather than raw fidelity. Some third-party coverage cited aggressive per-minute pricing for the API relative to premium competitors, which — if it holds up against real-world quality — would give small production teams and developers room to test far more variations before committing any clip to heavier post-production.
But there is a ceiling, literally. Grok Imagine Video 1.5 remains capped at 720p, while several rivals already ship 1080p. For social clips, rapid concepting, and short-form drafts, 720p is often good enough. For higher-end client work, resolution stays firmly on the adoption checklist, and it is the most obvious gap between Grok Imagine and the tools agencies reach for when the output has to survive a big screen.
It is also worth remembering the product line's baggage. The Grok Imagine family drew sharp criticism in 2025 over a "spicy" mode that generated non-consensual nude deepfakes, including of public figures — a controversy that still shadows xAI's positioning in the creative-tools market even as the underlying models improve.
What to watch
The single most telling signal for Grok Imagine's professional ambitions will be a 1080p option. Until that arrives, xAI is essentially betting that speed, synced audio, lower cost, and a smoother creator workflow are enough to win the volume end of the market while leaving premium client work to higher-resolution rivals.
Two other things bear watching. First, whether the synced-audio approach genuinely collapses the separate editing passes that slow short-form campaign work, or whether creators still find themselves doing cleanup. Second, how xAI manages usage limits going forward: the Voice Mode adjustments show a company actively tuning what subscribers get, and image-and-video caps could move the same way if demand outruns compute. In a market where Google, ByteDance, and others are shipping updates on a near-monthly cadence, the pace of iteration — not any single release — is what will decide where Grok Imagine lands.
---
Sources: [xAI — Grok Imagine Video 1.5](https://x.ai/news/grok-imagine-video-1-5); [WinBuzzer](https://winbuzzer.com/2026/06/18/xai-speeds-up-grok-imagine-video-15-for-ai-video-creators-xcxwbn/); [PiunikaWeb](https://piunikaweb.com/2026/05/13/xai-throttles-grok-paid-users/).
"These are our best image-to-video models yet: better motion, better physics, better audio, at the fastest speeds."- xAI, Company statement on Grok Imagine Video 1.5