Maya benchmarks every model release so you don't have to — numbers first, hype never.
Inspired by this story? Turn the idea into your own AI art in seconds — free to start, no card required.
Start creating freePick a companion and get their take on this story
Google has announced Gemini Omni 1.2 Flash, a new multimodal AI model offering native 15-second video generation and editing capabilities, alongside a 30% price reduction for its Flash API, making longer video content creation more accessible and cost-effective for AI artists and developers.
The headline feature of Gemini Omni 1.2 Flash is its native support for generating and editing video clips up to 15 seconds in length. This is a crucial development for AI video creators, who often grapple with limitations on clip duration and the fragmented workflows required to stitch shorter segments together. The model's ability to handle longer sequences natively should streamline the creative process, allowing for more coherent narrative arcs within a single generation. Previously, achieving this length often involved intricate prompting and post-processing, adding significant overhead to a creator's workflow.
Google states that the model also brings improved temporal consistency, a persistent challenge in AI video generation where objects or characters can change appearance erratically between frames. While independent tests will confirm the extent of this improvement, better consistency means less manual cleanup and potentially higher-quality outputs directly from the model, a direct benefit for anyone generating animated characters or dynamic scenes.
Accompanying the new capabilities, Google is cutting the price of its Flash API by 30%. This makes the model a more competitive option for creators who are sensitive to compute costs, especially when experimenting with longer video generations or iterating on prompts. For those using AI video tools for commercial projects, this cost reduction could significantly impact project budgets and profitability. The lower barrier to entry could also encourage broader adoption among independent artists and small studios looking to leverage AI for video production without prohibitive expenses. More affordable API access could also mean creators can iterate more frequently, refining their prompts and output without constantly watching their budget, a key factor in creative exploration. You can explore the current pricing structures for various AI models to understand the landscape of options available for your projects on our /pricing page.
Gemini Omni 1.2 Flash also features enhanced multimodal understanding. This means the model is designed to better interpret complex prompts that combine text, images, and now video inputs. For AI artists, this translates to more nuanced control over their generations. Instead of relying solely on text descriptions, creators can provide reference images or even short video clips to guide the AI, leading to outputs that more closely match their creative vision. This is particularly useful for maintaining specific visual styles, character designs, or motion patterns across different generated sequences.
This update follows other recent advancements in AI video, such as OpenAI's integration of GIF generation into ChatGPT with Images 2.6, allowing for direct animation creation from text prompts. As models like Gemini Omni 1.2 Flash continue to evolve, the ability to generate longer, more consistent video content at a lower cost will likely push the boundaries of what's possible for AI-powered visual storytelling. Creators interested in leveraging these new capabilities can explore various models and their features within our /catalog of AI generation tools.
