A video labelled "Claude Next Model" has been shared on X, according to @chetaslua, who says two upcoming image outputs will look as if they were "shot on camera but purely coded". The post is dated 11 October 2026. Anthropic hasn't announced a model called Claude Next.

What it shows

The post is brief. It reads "Claude Next Model :)", then declares "its over for image/video gen". The poster tells readers to wait for "two image gen output" with that camera-shot look, and closes with "never thought ai will get replaced by ai".

A video of about 51 seconds is attached. The text names no model version, product or tool, and shows no prompt or settings. "Claude Next" is the poster's own label, not an Anthropic name.

Video: Video attached to the X post by @chetaslua, which he labels 'Claude Next Model'; its content is not independently verified

Load the post to view it as published on X. X may receive connection data.

View the original post on X ↗

What it could mean

Our read: "purely coded" most plausibly describes a model writing code that draws a scene when it runs, rather than producing pixels directly the way image and video generators do. If so, this would be a story about coding ability, and "over for image/video gen" is the poster's opinion, not a finding. The post offers nothing to compare the output against existing image or video tools.

Anthropic has had other unconfirmed Claude model chatter recently, such as a Claude Code report about a rolling Fable selection and a Fable 5.5 URL that returned a 404. Nothing in this post ties "Claude Next" to either.

Three things would change the picture:

  • An Anthropic statement or model name.

  • Other testers reproducing the same kind of output.

  • The poster sharing the prompts and the generated code.

Status: unconfirmed. We'll update this page when Anthropic confirms or denies it.

Sources