ElevenCreative Chimera turns an image and a script into a talking character
The new tool targets short-form video creators. For now our information comes from a secondary source, as does the announced arrival of CapCut in ChatGPT.
Author
Articles signed “The Newsroom” are written and reviewed collectively by the editorial team. Every piece cites its primary sources and explicitly flags secondary sources and unconfirmed information.
How we work: editorial policy and methods. Write to us: contact.
The new tool targets short-form video creators. For now our information comes from a secondary source, as does the announced arrival of CapCut in ChatGPT.
On the same day, two flagship tools show the two sides of development agents: more capabilities on one hand, more guardrails on the other.
Using the standard protocol, the organization that runs the test gets a score well below OpenAI's, at a cost of more than $26,000.
Scored at 55 by Artificial Analysis, the model combines a top-tier ranking with aggressive pricing. Its weights are promised by November 4.
Announced at the end of September, Google's new model has been open to everyone since October 13. Another variant is reportedly being tested internally, without official confirmation.
The new video model is billed at 2.2 credits per second in 1080p. The real savings remain significant, but well below the marketing claim.
The new version of the video generator focuses on storytelling: longer shots and control through successive keyframes.
Three cents per 1K image and six reference images for Krea 3, versus about $0.034 and fourteen references for the Google DeepMind model.
Black Forest Labs releases a compact 5.6-billion-parameter model, free for commercial use for companies with under $5 million in revenue.
Weights, code and a permissive license: MOSS-Image comes in two versions, Base and SFT. Its strong layout ranking is, for now, self-reported.