What's new in Vidorena: Grok Imagine, Wan 2.5 and faster renders

A shorter update than our usual guides, but three changes worth knowing about if you have been away from the studio for a few weeks. New models, a speed pass, and a note on what is coming.

Grok Imagine, added

xAI's fast, expressive image and video model is now live across both the creator and business sides of the studio, listed alongside every other model on the Models page. It tends to get routed to shots that call for a bolder, more stylized look than the photorealistic default, and it is noticeably quick, which makes it a good fit for drafting several directions before committing a final render to a heavier model.

Tongyi's Wan 2.5

An image-to-video model with embedded audio, tuned for product shots. The useful part is not the video quality alone, it is that the audio is generated alongside the motion rather than layered on afterward, so the sound of a lid clicking shut or fabric moving actually lines up with the frame it happens in instead of being approximated later.

Render speed, studio-wide

Nothing user-facing changed here except the wait. Typical render times dropped across the board this month from routing and caching work on our end, most noticeably on multi-shot sequences where the orchestration layer can now start downstream renders before every upstream one has fully finished, instead of waiting on the whole plan to resolve first.

What's next

A few labs have flagged updated tiers landing in the coming weeks. We do not publish specifics until they are actually live in the studio, since release dates from labs move more often than not, but expect another of these notes before the month is out.