A leaderboard built around jobs
As reported by VentureBeat, OpenArt’s blind, task-based Arena ranks generative AI models across specific creative assignments instead of treating image and video generation as a single contest.
ByteDance’s Seedance 2.5 took the top position overall for video. It also led in film, motion design and lip sync, categories that test more than the ability to produce attractive frames. These tasks require models to preserve movement, timing and visual continuity, qualities that matter when an output must function as an advertisement, a music sequence or a talking character.
Alibaba’s Wan 3.0 narrowly won the video editing category. That distinction is important. A model that excels at generating a scene may not be the best tool for changing an existing clip, extending a shot or making a targeted correction. In a production environment, editing often saves more time than starting from scratch.
The image rankings were similarly divided. OpenAI’s GPT Image 2 led in graphic design and image editing, while Seedream 5.0 Pro won film-oriented imagery, e-commerce and the overall image board. For a creative team, the result resembles a toolbox rather than a throne. One model might produce a clean product listing, another a cinematic concept frame, and a third a precise revision to existing artwork.
Procurement with a measurement problem
That fragmentation could change how companies buy AI access. Instead of selecting one general-purpose provider, marketing departments may route each assignment to a specialized model, much as a studio chooses different cameras, lenses and postproduction tools.
Yet the Arena’s authority depends on details OpenArt has not fully disclosed. The company has not published final judge counts, prompt counts or total vote numbers, making it difficult to assess the statistical weight of each result. Blind testing can reduce brand bias, but it does not by itself reveal how representative the tasks or voters were.
The rankings are therefore useful as a map of emerging strengths, not a final verdict. As AI becomes embedded in everyday creative work, transparent evaluation will matter almost as much as visual quality. Teams will need to know not only which model wins, but why, under what conditions and with whose judgment.
Maya Lindqvist is not a person. No notebook, no deadlines, no face behind the name — just a byline this newsroom publishes under. Here is the production line underneath it, because a name beside a portrait reads like a journalist, and this one is not one.
The models. Writing: gpt-5.6-luna and qwen3-max. Out on the live web: gpt-5.6-luna and gpt-5.6-terra. Pictures: gpt-image-1 and gpt-image-1-mini. Swap one in the newsroom and this line swaps with it — it is read off the machines, not typed here.
How a story is made
- Research. The searching model reads around the story, pointed at primary sources — the filing, the post, the repository — rather than at somebody else's write-up of them.
- Writing. The writing model drafts it against what was found, at Maya Lindqvist's usual length and in Maya Lindqvist's usual register.
- The loop. A reviewer reads the draft and sends it back with notes. Then reads it again. A piece can go round several times before it leaves the building.
- Enrichment. A quotation has to appear word for word on the page it is taken from. A chart may only use figures that appear in the source it cites. Whatever fails is dropped, and the reason is kept.
- Fact check. A last pass hunts for claims the article makes and its sources do not.
- A human stop. Sensitive subjects are held for a person to read before publication, and a person can kill any of it at any point.
If that sounds less like a newsroom and more like a factory: quite. It is called Press Factory.
This article was generated using AI and published automatically without human pre-publication review.
How this article was made
The article was produced by the Grandmonts Media News Engine using automated research, drafting and verification workflows. No human editor reviewed the article before publication. Grandmonts Media remains responsible for the published content. Errors can be reported at office@grandmonts.cz.