Model release

Gemini Omni Flash 1.1 Reaches General Availability

Google released gemini-omni-1.1-flash as generally available on August 27, 2026. The update adds resolution controls for video generation and editing tasks.

Released Published 3 min readModel evidence ↗
Gemini Omni Flash Preview release illustration — Google

General Availability Announcement

Google officially released gemini-omni-1.1-flash into general availability on August 27, 2026. This model serves as the GA version of the fast, conversational video generation and editing system previously known as Gemini Omni Flash Preview. The source confirms that this specific release marks the transition from the preview status to a stable general availability state for the gemini-omni-1.1-flash identifier.

New Resolution Capabilities

The primary documented change in this release is the addition of a resolution parameter within the videoconfig object. This parameter supports four distinct output resolutions: 360p, 720p, 1080p, and 4K. According to the official changelog, 720p is set as the default resolution. For higher resolutions, specifically 1080p and 4K, the source states that these outputs are generated using upscaling techniques. No other architectural changes or performance metrics were detailed in the provided facts regarding how this upscaling process functions internally.

  • Supported resolutions include 360p, 720p (default), 1080p, and 4K.
  • 1080p and 4K outputs utilize upscaling.
  • The resolution parameter is located in the videoconfig structure.

Endpoint Deprecation and Evaluation Steps

The existing gemini-omni-flash-preview endpoint will be deprecated on September 30, 2026. Engineers should plan to migrate any current integrations from the preview endpoint to the new gemini-omni-1.1-flash endpoint before this date. To evaluate the new model, developers can test the resolution parameter by generating videos at each supported tier. Measure the output file sizes and visual fidelity differences between the native 720p output and the upscaled 1080p/4K results. Compare the processing time for different resolutions to validate if the 'fast' designation holds across all quality levels. Note that open weights are not confirmed for this model, and pricing information was not provided in the source facts.

Primary sources

QUICK ANSWERS

Frequently asked questions

When does the old preview endpoint stop working?

The gemini-omni-flash-preview endpoint is scheduled for deprecation on September 30, 2026. Users should transition to the gemini-omni-1.1-flash endpoint prior to this date.

How are 1080p and 4K videos generated?

According to the source, 1080p and 4K outputs are generated using upscaling. The default resolution remains 720p.

More from the journal

DeepSeek V4 Flash Vision Exp: Preview Release Facts →

On August 21, 2026, DeepSeek announced the experimental multimodal model DeepSeek-V4-Flash-Vision-Exp. Available via the DeepSeek API, it accepts text and image inputs to produce text outputs. Official documentation lists specific benchmark scores for various tasks, while context window limits and pricing remain unspecified in the source material.

GPT-6 Astra Release: Capabilities and API Limits →

On September 3, 2026, OpenAI released GPT-6 Astra via the v1/responses and v1/chat/completions APIs. The source describes it as built for reasoning, coding, computer use, research, and document creation. Engineers must note that the model does not support the 'none' reasoning effort level, custom temperature or topp values, or log probabilities.