Silicon Valley startup Luma AI has launched Ray3. It says it is the first video model to have the ability to reason.
Adobe and leading advertising and creative agencies Monks UK, Galeria and Strawberry Frog are the first customers of Ray3 available on the Dream Machine Platform.
Generation models can criticize themselves and create on-screen notes about revisions and improvements. Unlike many current models, production work can also be saved and taken to a stop. “Draft Mode” allows for more trial and error at a faster rate, and renders with priority results with 4K high dynamic range.
Luma says the output of a system that is blessed with the power to reason is “a character that appears to be consistent, scenes that unfold naturally over time, and videos of each person behaviour.”
Luma also states that the Ray3 is the first model to create videos in the true 10, 12, and 16-bit HDR ACES2065-1 EXR format, making it suitable for high-end movies and advertising projects.
Supporters Luma, including Nvidia and venture capital firm Andreessen Horowitz, opened their LA studio earlier this year, making its presence in New York.
“Ray3 is the first step in building intelligence for creative work,” said Amit Jain, CEO and co-founder of Luma AI. “Creative work is one of the most intellectually challenging things humans do, but up until now, many of the AI that creatives can use have slowed down what is possible with coding and analysis in language models. Generation models are similar to slot machines.
Adobe and Luma AI have partnered to bring Ray3 directly into the Adobe Firefly app.
