Gemini Omni Flash video model goes GA at about $0.10 per second

Google released gemini-omni-1.1-flash on the Gemini API on August 27, 2026. The changelog calls it the GA (generally available) version of Gemini Omni Flash, Google’s fast model for making and editing video by conversation. It replaces gemini-omni-flash-preview, which Google deprecated on September 30, 2026.

What it does

• Generate and edit video from text, images and video in one request. Video references are limited to 3 clips of up to 3 seconds each.
• Extend a clip with a 3 to 10 second continuation, up to 40 seconds in total. Extension only adds to the end.
• Interpolate between two images, as a first and last frame.
• Resolutions: 360p, 720p (the default), 1080p and 4K. The 1080p and 4K outputs are upscaled.
• Format: 16:9 by default, or 9:16 for vertical video.
• Audio is generated by default from your prompt.
• Watermark: every video carries an invisible SynthID watermark.

Editing or extending videos you upload isn’t available in the European Economic Area, Switzerland or the UK. Google’s models overview page still files the model under its preview section, although the changelog calls it GA.

What it costs

Google bills Omni Flash by the token, with no free tier:

Item	Price per 1M tokens
Input (text, image, video, audio)	$1.50
Text output	$9.00
Video output	$17.50

A second of 720p video counts as 5,792 output tokens, which Google says comes to about $0.10 per second. The pricing page gives no per-second figure for 360p, 1080p or 4K.

Older Veo models are gone

Separately, Google shut down veo-2.0-generate-001, veo-3.0-generate-001 and veo-3.0-fast-generate-001 on the Gemini API on June 30, 2026, and pointed users to the Veo 3.1 models. The chart above compares Omni Flash with Veo 3.1’s per-second prices at 720p.

Since 1080p and 4K are upscaled, test whether a 720p clip at about $0.10 a second meets your needs before paying for a higher setting.