Kling 4.0 is now available in Luma, which pitches the model for fast action and bigger camera moves. Luma's demo combines several creative models, including Kling 4.0.
Scan real objects and scenes with a phone camera into photorealistic 3D models using NeRF, or generate 3D assets and video with Dream Machine.
About Luma AI
Luma AI captures the world in lifelike 3D using Neural Radiance Field (NeRF) technology. Scan real-world objects and scenes with just your phone camera to create photorealistic 3D models you can view, share, and embed anywhere. Luma's Dream Machine also generates 3D assets and videos from text or image prompts.
Whether you're a product designer creating digital twins, a real estate agent offering virtual walkthroughs, or a creative experimenting with 3D content, Luma makes 3D capture and generation accessible to everyone.
Pricing model: Free
Features
- NeRF Technology
- Interactive Scenes
- Dream Machine
FAQs
Common questions about Luma AI
What do I need to create 3D scans with Luma AI?
Just your phone camera — Luma uses NeRF technology to turn simple phone captures into photorealistic 3D models.
What is Luma's Dream Machine?
Dream Machine is Luma's AI feature that generates 3D assets and videos from text or image prompts, no scanning required.
Luma’s Style Match restyles product shots from 1–4 references
Luma AI showed Style Match, which restyles your image using one to four reference images.
Luma says it keeps the product and composition in place while applying the references’ art direction, so you can try different looks on a product shot without rebuilding the layout.
Luma Shot Coverage turns one video take into new camera angles
Luma’s Shot Coverage turns one short video take into new moving camera angles you can cut together.
Choose up to all three coverage types: performance-led angles follow the actors, action-led angles focus on hands and props, and spatial-led angles show the room. Each type comes back as its own take, with options for output resolution and sound.
Live Text Extraction in Layers
Luma Layers can now pull baked-in text out of an image as editable live text, matching it to the closest font in its library.
Select a layer and hit Extract. You can then retype, swap the font, or recompose the design without regenerating the image. The feature is powered by Uni-1.
Read the original announcement →
GPT-Image 2.5 in Luma Agents
GPT-Image 2.5 is live in Luma Agents, with Flare and Sunburst modes.
Luma positions Flare for speed and volume, and Sunburst for precise edits and maximum fidelity. With Sunburst, you can bring a reference image, change specific details while keeping the rest, then take the result into video.
Read the original announcement →
Seedance 2.5 on Luma
Luma added Seedance 2.5 for up to 30 seconds of multi-shot, high-definition video in a single generation.
You can guide it with up to 50 references across text, image, video, and audio. Luma Agents lets you control scenes, camera angles, and pacing.
Layers in Luma Agents
Luma Agents now has Layers. Ask the agent to extract layers from an image you've generated in Luma or uploaded, then keep editing in the same chat.
You can ask it to swap a product, restyle a background, or rewrite a headline. Luma says Layers lets you change one element while preserving everything else. It runs on Uni-1.
Seedance 2.0 Mini in Luma
Seedance 2.0 Mini is now available in Luma AI. Luma pitches it as a fast video model with a lighter footprint for storyboarding concepts and testing different directions.
Luma Connectors
Luma AI's new Connectors let you link Airtable, Dropbox, and Google Drive, then pull files from those accounts into any Luma board on demand.
Read the original announcement.
Seedance 2.0 Joins Luma Agents
Luma Agents can now generate with Seedance 2.0. Choose it for a project, review the output, and keep working in the same agent workflow.
Loading more posts
Loading mentions…
Alternatives to Luma AI
Other products to explore and follow.
Team
The people behind Luma AI
Amit Jain
Co-founder and CEO