MUMBAI: Luma today announced the launch of Luma Agents, a new class of AI collaborators capable of performing end-to-end creative tasks across text, images, video, and audio. Designed for agencies, marketing teams, studios, and enterprise organizations looking to scale their creative output without sacrificing quality, Luma Agent maintains complete context from initial brief to final delivery, orchestrating tools, models, and iterations within one unified system.
“Creative work has never lacked ambition. It has never lacked the ability to execute. Creative teams don’t need to spend time adjusting tools; they should spend time creating. Agents are not shortcuts. They are collaborators who maintain context, orchestrate execution, and move projects forward so teams can focus on preference, direction, and strategy.” said Amit Jain, Luma Co-Founder and CEO
For the past few years, most AI systems have been assembled by chaining together separate linguistic, visual, video, and inference models and stitching their outputs together through orchestration layers. While powerful on their own, these systems fragment context and require increasingly complex workflows to produce reliable creative results.
Mr. Luma believes that intelligence should not be assembled in pieces. It must be built as one coherent system.
Creative agency that makes you prolific
Luma agents replace fragmented, multi-model workflows with coordinated execution built on unified reasoning. Rather than switching between disconnected tools or rebuilding context at each step, teams collaborate with agents to:
- We execute projects end-to-end, from planning to production to delivery.
- Maintain shared context across text, images, video, and audio
- Working in multiple creative directions in parallel
- Evaluate and adjust output rather than producing one-shot results
- Integrates into enterprise tools and production systems via API
Agents operate within a collaborative multiplayer environment where humans direct creative intent and agents handle orchestration, routing, and execution. The result is more output, consistency, and higher creative speed.
Expanding worldwide
Luma agents are already integrated across global agency operations.
Publicis Groupe Middle East and Serviceplan Group deploy Luma Agent across strategy, creative development and production workflows to increase throughput while maintaining brand consistency across markets.
“Luma is now part of our broader House of AI ecosystem and integrated directly into our creative workflows, allowing our teams in over 20 countries to work together more smoothly and develop great work faster. For our clients, this means we can deliver high-quality creative deliverables faster and more efficiently without compromising craftsmanship.” said said Alexander Schil, Global CCO of Serviceplan Group.
Built on unified intelligence
Luma agents are built on Unified Intelligence, a new model architecture designed to go beyond the industry’s common approach of assembling intelligence piecemeal. Instead of chaining together separate linguistic, visual, and generative models, Unified Intelligence trains a single multimodal inference system that can understand and produce different formats within the same architecture.
For the past few years, most AI systems have been assembled as pipelines. That is, one model writes the text, another generates the image, another processes the video, and an orchestration layer tries to stitch the output together. Although these systems are effective for narrow tasks, they fragment inference, lose context between steps, and require complex workflows to produce reliable results.
Unified Intelligence takes a different approach. Rather than connecting specialized models after the fact, train a single multimodal inference system that can understand and generate different formats within the same architecture.
Rather than separating thinking and creating, unified intelligence tightly couples reasoning and rendering, allowing systems to plan, imagine, and generate as part of one coherent cognitive process.
When human architects sketch buildings, they don’t just draw lines; they simultaneously simulate structure, light, spatial dynamics, and lived experience. Reasoning and imagination occur simultaneously. Unified Intelligence is built on the same principles.
The first model built on this architecture was the Uni-1.
Uni-1 is a decoder-specific autoregressive transformer that operates on a shared token space that interleaves language and image tokens, allowing both modalities to act as first-class inputs and outputs within the same sequence. This design allows the model to reason with the language while imagining and rendering pixel-by-pixel within the same forward pass.
Uni-1 can plan, visualize, and produce creative artifacts as part of a single, coherent reasoning process, rather than incrementally producing output across disconnected systems. The result is a close connection between thinking and creation, creating a foundation much closer to the workings of human intelligence.
Built on this unified architecture, Luma agents can orchestrate complex creative workflows that previously required multiple tools and manual orchestration. They can:
- Adjust between leading AI models such as Ray3.14, Veo 3, Sora 2, Kling 2.6, Nano Banana Pro, Seedream, GPT Image 1.5, Celebrities, and more.
- Automatically select tasks and route them to the best model or feature for each step.
- Maintain persistent context across assets, collaborators, and creative iterations
- Improve results through iterative self-criticism, evaluate and refine output
Together, these capabilities allow Luma agents to function not as isolated generation tools, but as collaborative AI creatives capable of performing end-to-end creative work.
“Intelligence should not be fragmented by modality. Integrated systems reason holistically. When the same model can think, imagine, and render, we move closer to intelligence operating consistently throughout the creative process.” added Jainism.
Enterprise-ready design
Luma Agent is designed for enterprise environments where intellectual property protection, compliance, and operational scale are important. Key enterprise safeguards include full IP ownership retained by customers, automated content review to reduce copyright risk, legal documentation to prove human involvement, human review workflows required prior to publication, and cloud-based infrastructure with enterprise-grade guardrails.
Link –
