Luma has released the AI video editor Ray3 Modify - ForkLog: cryptocurrencies, AI, singularity, future

# Luma Releases AI Video Editor Ray3 Modify

The AI-driven video and 3D modeling company Luma has launched a new neural network, Ray3 Modify.

The tool allows users to modify existing videos using reference images of characters while preserving the acting from the original footage.

Users can specify start and end frames for the model to generate intermediate content.

The company states that Ray3 Modify addresses the challenge of maintaining character features during editing or effect generation. LLMs better follow the original video, preserving natural movements, timing, gaze direction, and emotional expression. This enables studios to use live actors in creative videos.

With this new solution, users can set a reference character to transform the video, changing the person’s appearance. Information about costumes, physical resemblance, and identity is maintained throughout the shoot.

“Generative video models are incredibly expressive, but difficult to control. Ray3 Modify combines richness and realism, giving creative professionals full control. They can record performances on camera and then instantly modify them: transform the location, costumes, or even ‘reshoot’ scenes with AI without physical re-shoots,” — said Amit Jain, co-founder and CEO of Luma AI.

The new model is available on the Dream Machine platform. The release coincided with $900 million in funding raised in a round led by Saudi-based Humain. Existing investors in the project, including a16z, Amplify Partners, and Matrix Partners, also participated.

The startup plans to build a 2 GW AI cluster in Saudi Arabia in partnership with Humain.

Recall that in May, American tech companies signed an agreement with a company from an Arab country to scale AI development in the region.

In December, AI startup Runway released a new video model, Gen 4.5, which outperforms competitors in independent testing.