Gemini Omni 1.1 Flash
gemini google multimodal
| Source: HN | Original article
Google's Gemini Omni 1.1 Flash is now rolling out, adding a suite of creative controls and generative video features for developers.
Google DeepMind has moved its next‑generation video model, Gemini Omni 1.1 Flash, out of pre‑general availability and into full rollout on Google Cloud. The update follows a brief delay in the Pre‑GA phase, and the company now lists a suite of multimodal features, latency improvements and tighter integration details for developers.
Gemini Omni 1.1 Flash expands the creative toolbox that Google first unveiled in late August. In addition to the 4K upscaling and scene‑extension capabilities highlighted then, the new version adds first‑and‑last‑frame controls, longer scene extensions and a “cheaper low‑cost” tier for budget‑sensitive workloads. Developers can generate videos with native audio from text, images or keyframes, and the model supports image‑to‑video, keyframe interpolation and reference‑to‑video pipelines at up to 4K resolution.
The significance lies in the combination of speed, quality and control. By lowering latency and offering granular editing handles, Gemini Omni 1.1 Flash makes high‑fidelity AI video production more accessible to a broader range of creators, from advertising agencies to independent developers building custom media tools. Native audio generation further narrows the gap between AI‑generated content and traditional studio workflows, potentially reshaping how video is produced at scale.
What to watch next includes how quickly developers adopt the new controls and whether Google’s pricing tiers spur wider experimentation. Performance benchmarks against competing models will likely surface as early adopters publish results. Additionally, Google may announce further enhancements—such as deeper integration with its broader Gemini suite or expanded cloud‑native tooling—that could cement the platform’s role in the rapidly evolving generative video market. As we reported on 27 August, the model already promised studio‑quality output; the current rollout confirms that promise is now available for production use.
Sources
Back to AIPULSEN