Google has released Gemini Omni 1.1 Flash (gemini-omni-1.1-flash), an updated multimodal video generation and editing model available through the Gemini API and enterprise platforms. The update enables stateful, multi-turn video editing via the Interactions API, allowing developers to extend generated videos in 10-second increments up to 40 seconds total. Omni 1.1 Flash reads up to 10 seconds of prior visual context to maintain continuity across scene extensions.
The model introduces dedicated framing control via prompt tags (<FIRST_FRAME>, <LAST_FRAME>, and image/video reference tags) to guide camera trajectories, character consistency, and seamless looping. To lower production rendering costs, Google introduced a 360p draft resolution mode priced at one-third the cost of 720p, paired with a 4K upscaling pipeline. Standard API pricing is set at $1.50 per million input tokens and $17.50 per million video output tokens.
Google confirmed native integration with third-party creative platforms, including Adobe, Figma Weave, and Runway. All generated output incorporates invisible SynthID digital watermarking for provenance verification, though the release currently omits features like temperature control, audio reference inputs, and negative prompts.
Why it matters
Provides enterprise developers with cost-effective draft-and-upscale workflows for API-driven commercial video generation.
Enables precise camera and character control using frame-pinning primitives rather than relying solely on text prompts.
Strengthens native multimodal API capabilities for enterprise workflows across creative tools like Adobe, Figma, and Runway.
Source: marktechpost.com
