SlopTV

news

VideoGen unveils Model 1 with a ten-minute ceiling and no published price, benchmark or third-party number

The October 20 launch release says Model 1 should cost more than 50% less than comparable models, then states that final pricing, official benchmarks and third-party validation are still underway.

By Daniel Ochoa ·

VideoGen announced Model 1 on October 20, 2026, an AI video model that assembles multi-shot videos up to ten minutes long from one prompt or up to 16 reference images. The company expects it to cost more than 50% less than comparable models based on internal validation, with pricing, benchmarks and third-party validation still pending.

VideoGen announced its first in-house video model on October 20, 2026, and the announcement contains a ten-minute duration ceiling, a projected cost more than 50% below comparable models, and an explicit admission that none of that has been priced or independently measured yet. The release states that final pricing, official benchmarks and third-party validation are underway, which leaves the quality bar for a ten-minute generated video entirely undefined on launch day.

Model 1 takes a written brief plus optional visual references, up to 16 images, along with a target duration and aspect ratio. A planning layer the company calls a thinking harness writes the screenplay, dialogue, shots and pacing, carrying context between scenes so characters, locations and audio stay consistent. Both text-to-video and image-to-video are supported, and the generation architecture uses joint audio-video attention operating in a compressed latent space that is decoded into frames and sound.

Introducing VideoGen Model 1 · YouTube

The ten minutes is an assembly length, not one continuous generation

Longer videos are assembled from multiple generated shots. The production pipeline checks outputs for problems including incomplete dialogue and visual inconsistencies, retries selected renders inside a set budget, then trims and joins the chosen clips. That is a meaningfully different artifact from a single ten-minute sample, and it is the same structure behind other long-duration claims in this market, including Higgsfield's ten-minute figure, which also arrived without a sample attached. VideoGen says example videos live at videogen.io/model-1 and that a technical whitepaper covering its research approach, experiments and findings is still to come.

A 50% cost reduction measured internally, against models the release never names

The cost line is specific about its own limits: based on internal validation, VideoGen expects Model 1 to cost more than 50% less than comparable models. No comparison set is named, no per-second or per-clip figure is given, and the company's own text puts final pricing in the future tense. A percentage cut without a baseline cannot be checked by anyone outside the company.

What one 1080p, 10-second clip of Seedance 2.5 costs
Lumeta$5.41
Higgsfield$5.51
BytePlus ModelArk$5.69
Runway$6.80
Runway API$6.80
Kie.ai$7.90
Higgsfield API$7.96
Magnific$9.68
fal.ai$11.64
Cheapest published price on each platform's best plan, read by our price radar on 21 Oct 2026. Live comparison

For scale, that chart shows what a 10-second 1080p clip from Seedance 2.5 costs across the platforms on our price radar, which is the kind of published number a 50% claim would need to be measured against.

One speed figure did make it into the release with its caveat intact. In an internal experiment, an eight-step adapter cut average clip-render time by 58.2% compared with the same system's 28-step baseline, and VideoGen writes that this measures an experimental rendering configuration rather than a guaranteed speed improvement across all Model 1 outputs. Attaching that qualifier is more disclosure than most launch numbers get, and it also means the figure describes a configuration that may not be the one users receive.

A $1 million credit pool, with selection criteria standing in for a price

Alongside the model, VideoGen opened applications on October 20 for $1 million in free credits aimed at creators, filmmakers and studios, including students and amateurs. Credits are to be distributed through the end of 2026, with individual allocations based on each project's scope and scale, and selection weighing both creative work and a viable distribution plan. Model 1 ships with guardrails intended to block adult and illegal content, and outputs carry invisible watermarks marking their synthetic origin.

The company said in September that it had passed 5 million users across more than 190 countries and raised a $3.3 million seed round, so Model 1 is its first owned model after years of routing customers through third-party generators.

Sources: VideoGen via GlobeNewswire, VideoGen seed announcement.

Cite this

Free to cite and reuse with a link back. Data is updated as new runs and prices come in, so include the date.

SlopTV. (2026). VideoGen unveils Model 1 with a ten-minute ceiling and no published price, benchmark or third-party number. Retrieved October 21, 2026, from https://sloptv.co/news/videogen-model-1-ten-minute-claim-pricing-pending
<a href="https://sloptv.co/news/videogen-model-1-ten-minute-claim-pricing-pending">VideoGen unveils Model 1 with a ten-minute ceiling and no published price, benchmark or third-party number</a> (SlopTV)

Daniel Ochoa

Daniel Ochoa: Covers model launches, shutdowns and pricing changes as they happen. Reads deprecation notices for a living so you do not have to.