Stages AI adds 96-shot storyboard pipeline for Pro users
Stages AI posts show Pro users can turn a script into a 96-shot storyboard in about 10 minutes and send shots into VIDX and a timeline editor in two clicks. The workflow compresses previsualization and assembly into one pipeline, but the throughput claims come from creator demos rather than a formal product spec.

TL;DR
- dustinhollywood's first demo says Stages AI Pro can turn a script into a 96-shot storyboard in about 10 minutes, then hand those shots to VIDX for animation in another 25 minutes when concurrency is capped at 4x.
- In dustinhollywood's follow-up thread, the same workflow expands from storyboarding to full shot generation, with claims that a 120-minute feature can be assembled from script to shot film in a day through linked generation and editing tools.
- dustinhollywood's VIDX clip shows the handoff path clearly: storyboard to "director mode" in VIDX, then into a timeline editor in two clicks.
- Access is already being pitched to paying users, because dustinhollywood's sign-up post points directly to the Pro waitlist at Stages Pro.
- The big caveat sits in the evidence itself: the speed and scale numbers come from creator demos like the 96-shot timing post and the feature-length claim, not from a formal product spec or benchmark page.
You can already join the Pro sign-up page, browse the broader Stages AI site, and the most interesting reveal is not just the 96-shot count. It is the linked pipeline in the VIDX handoff demo, plus the editing clip showing music and SFX generation tracking the footage being cut. A later iPad post suggests the same workspace is usable from a tablet, and the June 1 teaser sets up a live public demo rather than another edited sizzle reel.
96-shot storyboards
Stages AI's clearest claim is simple: a script becomes a 96-shot storyboard in about 10 minutes. The same post says animating those shots inside VIDX took another 25 minutes at a 4x concurrency rate, and argues that higher concurrency would push the system toward much longer boards in much less time.
That matters as a previsualization story, not just a generation story. The post frames the output as something close to instant script visualization, with the creator saying "every script that exists can now be visualized in under a few hours throttled" in the original timing thread.
VIDX and timeline editor
The pipeline has three concrete steps in the tweets:
- Generate every shot from a script or brief.
- Send every shot into VIDX in exact order, with metadata and prompts attached, per the autonomous pipeline thread.
- Drop the result into a timeline editor in one more click, as shown in the storyboard-to-editor clip.
The strongest product reveal is the metadata handoff. dustinhollywood's thread says prompts, order, and shot data survive the jump from generation into editing, which is the part most AI video tools still leave to manual cleanup.
Any style, plus sound
Styles and post work are bundled into the same pitch. The first demo says the system already has a style reference layer, while a separate post reduces the message to two words, "Any style," attached to another demo clip.
The editing side goes further than picture. In the sound-generation post, dustinhollywood says Stages can track the content being edited and generate SFX and music against that timeline, which pushes the tool closer to an assembly environment than a one-off video model front end.
Throughput claims
Most of the huge numbers come from one creator account, and they are framed as practical ceiling tests rather than a published spec. The follow-up thread says the system can generate an entire 120-minute film in three clicks and hold continuity across the whole run in less than a day, while the earlier post estimates that 40x concurrency could storyboard roughly 1,200 shots in about an hour.
Those are eye-catching claims, but they are still demo claims. There is no linked benchmark page, changelog, or documentation in the evidence that pins down maximum shot counts, concurrency tiers, supported models, or continuity controls.
Access and surfaces
The availability detail is narrow but concrete. The original demo says the feature is available for Pro users, and the sign-up post links directly to Stages Pro.
A later post adds one new surface: dustinhollywood's iPad note says he was using Stages on an iPad for client work on a dome project. That is the first evidence in the thread that the workflow is not chained to a desktop setup.
June 1 live demo
The next concrete date is June 1. dustinhollywood's thread promises a live X demo of STAGES PRO, explicitly framed as a real-time showing with "no ambiguity" and without heavy motion-graphics editing.
If that stream happens as promised, it should answer the biggest open questions left by the current posts: how much of the pipeline is automated, how continuity is maintained across dozens of shots, and what parts still depend on a skilled human operator.