Palo Alto-based artificial intelligence company Luma AI has unveiled its latest video generation model, Ray2 Flash, which was launched last week. Described as a streamlined version of the previously released Ray 2 frontier model, Ray2 Flash is engineered to operate more quickly and cost-effectively while maintaining high-quality video generation capabilities. This new model is multimodal, accepting both text and image inputs, and is accessible through the Dream Machine platform introduced last year.
Luma AI Unveils Ray2 Flash Video Generation Model
An announcement on Luma AI’s official X account (formerly Twitter) detailed the launch of the Ray2 Flash model. It is currently integrated within the Dream Machine platform, with an application programming interface (API) slated for release in the near future.
The announcement touted that the Ray2 Flash model is three times faster and more affordable than its predecessor, the Ray2. Furthermore, it eliminates rate limits, allowing users to generate more videos without additional credits. However, Ray2 Flash is exclusively available to subscribers of Dream Machine.
Free-tier users receive an initial allocation of 400 one-time credits, but must subscribe for further access. Monthly memberships for the platform start at $6.99 (approximately Rs. 609), providing 3,200 monthly credits.
This new video generation model builds on the foundation laid by the frontier Ray2 model, which was launched in January. During its introduction, Luma AI highlighted that the frontier model is capable of producing realistic videos that exhibit natural motion and coherence, while also demonstrating a strong adherence to user prompts.
The Dream Machine platform, launched last year, is designed to facilitate both image and video creation across various styles, including cinema-grade, animations, and realistic graphics. However, users have noted some challenges, such as the model’s prompt adherence being less than satisfactory, particularly when dealing with complex prompts or multiple characters.
