Grok Imagine Video 1.5 Beats Sora, Veo, and Kling in Blind User Benchmarks at 86% Lower Cost

Grok Imagine Video 1.5 beat Sora 2, Veo 3.1, and Kling in blind user benchmarks at 86% lower cost, with native audio and dialogue in one pass.

Monday June 22, 2026 Source: x.ai
TL;DR — Quick Answer

xAI made Grok Imagine Video 1.5 generally available on June 22, 2026, and it beat OpenAI Sora 2, Google Veo 3.1, and Kuaishou Kling in blind user preference benchmarks while producing clips at 86% lower cost. It generates sound effects, ambience, and dialogue in the same pass as the video, and ships as grok-imagine-video-1.5 on the Imagine API with a faster 1.5 Fast model in the Grok apps.

Key Takeaways

Grok Imagine Video 1.5 Beats Sora, Veo, and Kling in Blind User Benchmarks at 86% Lower Cost — AI news article illustration

xAI made Grok Imagine Video 1.5 generally available on June 22, 2026, calling it the company’s best image-to-video model yet — and the numbers back it up. In blind user preference testing, Video 1.5 beat OpenAI’s Sora 2, Google’s Veo 3.1, and Kuaishou’s Kling, while costing 86% less per generated clip.

The Blind Benchmark Win

In head-to-head blind tests, users picked Grok Imagine Video 1.5 outputs over all three rivals without knowing which system produced any given clip. xAI attributes the win to improvements across the dimensions that matter for real creative work:

Audio That Lands on the Action

Most text-to-video models need a second pass, or a separate audio model, to add sound. Grok Imagine Video 1.5 generates sound effects, ambience, and dialogue in a single pass, with speech that is clearer and better synced to mouth movement. That removes an entire post-production step for short-form creators.

86% Lower Cost Per Clip

Cost is where Video 1.5 separates from the pack. At roughly 86% below Sora 2 per generated clip, a clip that costs a dollar on Sora runs about fourteen cents on Grok Imagine. For teams generating hundreds of test shots while storyboarding, that gap swings real production budgets.

Availability

The model ships as grok-imagine-video-1.5 on the xAI Imagine API, and a faster 1.5 Fast variant has rolled out on grok.com/imagine plus the iOS and Android apps. New workflow features include Projects for organizing work, multiple parallel agents, and library search.

What This Means

With native synchronized audio and an 86% price advantage, xAI is attacking the two things that keep video models out of production: sound and cost. The win in blind benchmarks just makes the price gap harder for the incumbents to ignore.

Frequently Asked Questions

Which AI video model won this benchmark?

Grok Imagine Video 1.5 from xAI, announced on June 22, 2026. In blind user testing it beat OpenAI Sora 2, Google Veo 3.1, and Kuaishou Kling.

How much cheaper is Grok Imagine Video 1.5 than Sora?

It costs roughly 86% less per generated clip. A clip that costs one dollar on Sora 2 runs about fourteen cents on Grok Imagine Video.

Does Grok Imagine Video 1.5 generate audio?

Yes. Sound effects, ambience, and dialogue are generated in the same pass as the video, with speech that is clearer and better synced to the action.

Where can I use Grok Imagine Video 1.5?

It is available on the xAI Imagine API, on grok.com/imagine, and in the Grok apps on iOS and Android, alongside new Projects, parallel agents, and library search features.

This article is based on the official announcement from x.ai . Read the original for full technical details.

Related Articles

Back to all news