Luma AI Launches Ray3: Revolutionizing Video Generation with Reasoning Capabilities, Supporting 16-Bit Color Depth

AIbase基地

Published inAI News · 4 min read · Sep 19, 2025

Artificial intelligence company Luma AI has recently launched its latest video generation model, Ray3, aiming to revolutionize the video creation process through its unique HDR (high dynamic range) capabilities. Ray3 claims to be the first AI model capable of generating studio-quality HDR videos, marking a significant step forward in video generation technology.

Seamless Integration of HDR with Professional Workflows

The most impressive feature of Ray3 is its support for high dynamic range video. The model can not only generate videos with 10-bit, 12-bit, or even 16-bit color depth but also export the results in professional EXR file formats, making it easy to integrate into film production, post-editing, and color grading workflows. In addition, Ray3 is capable of converting standard SDR videos into HDR, offering content creators more possibilities.

Innovative "Reasoning" Capabilities and Visual Control

Luma AI emphasizes that the core advantage of Ray3 lies in its unique "reasoning" capability. The model can think like humans, understand and follow complex instructions, and even self-evaluate the quality of its output. According to the company, Ray3 can review generated drafts and continuously iterate until it meets the set quality standards.

This reasoning ability has also led to innovative "visual control" features. Users are no longer limited to text prompts but can directly draw sketches on images to precisely control movement, objects, or camera framing in videos, providing unprecedented freedom for creation.

To balance cost and efficiency, Ray3 also introduces a "draft mode," allowing users to generate rough video previews at a lower cost and faster speed. These drafts can be upgraded to full 4K HDR "Hi-Fi" quality after secondary rendering.

A Leap in Detail and Fidelity

Luma AI emphasizes that Ray3 has made significant progress in fidelity and realism. The model has shown marked improvements in rendering crowd scenes, handling light interactions and reflections, simulating motion blur, and maintaining character consistency. Currently, the model is available for free to users through Luma AI's Dream Machine application.

Strategic New Growth Point! Map Giant AutoNavi Launches Robotaxi Business

AutoNavi officially enters the Robotaxi field, with a strategic goal targeting the global market. Internally, it is positioned as a strategic business, expected to become an important growth sector. This move aligns with the rising trend in the autonomous driving industry, and AutoNavi leverages its digital map advantages to expand its business boundaries.

AI Large Model Investment Competition Concludes! Alibaba Tongyi Qianwen Qwen3-Max Wins with a 22.32% Return Rate

The first AI large model investment competition organized by the US Nof1 institution concluded, with Alibaba's Tongyi Qianwen Qwen3-Max winning with a 22.32% return rate, demonstrating its leading capabilities in quantitative trading. The competition provided each of the six top models with $10,000 in initial funds to compete in a real trading environment on the Hyperliquid platform.

Kunlun Wanyi SkyReels V3 Model Launch! One-Stop Aggregation of Top AI Video Capabilities such as Sora2 and Veo3.1

The AI video creation platform SkyReels under Kunlun Wanyi is officially launched with a new version, introducing the V3 model and five core function upgrades, supporting both web and mobile ends. The platform highlights the 'one-stop' and 'multi-modal aggregation' features, integrating top global AI multi-modal models to provide a seamless creative experience.

Coca-Cola's New AI-Generated 'Coca-Cola Wagon' Ad Draws Criticism, but the Company Insists It's Faster and More Efficient

Coca-Cola once again used AI technology to create a holiday ad, launching the new 'Coca-Cola Wagon', but consumers criticized its visual effects as uncomfortable, finding it even more bizarre than last year's version. This year's ad features animal characters and rough animation, aiming to avoid the negative feedback from last year's issues with the roller coaster wheels and eerie faces, but it still failed to gain approval.

Baidu E-commerce Introduces Large Models to Reshape Risk Control Review Process, Achieving Win-Win for Merchants and Users!

"Baidu E-commerce Selection" brand uses large model technology to optimize risk control review, achieving full machine review, instant feedback, and high interpretability, solving the problems of low efficiency and slow response in traditional manual review, and enhancing e-commerce security and user experience.

The Controversy Surrounding Google's Gemma Model Reveals the Risks in the AI Development Lifecycle

Google's Gemma model sparked controversy after it spread false information about U.S. Senator Blackburn. It was accused of defamation. Google announced on October 31st that it would remove the Gemma3 model from the AI Studio platform to prevent misunderstandings. The model is now only available via API, and no longer accessible on AI Studio.