{"id":114590,"date":"2026-08-30T00:53:23","date_gmt":"2026-08-30T00:53:23","guid":{"rendered":"https:\/\/youzum.net\/google-ai-releases-gemini-omni-1-1-flash-40-second-scene-extension-first-last-frame-control-and-4k-upscaling\/"},"modified":"2026-08-30T00:53:23","modified_gmt":"2026-08-30T00:53:23","slug":"google-ai-releases-gemini-omni-1-1-flash-40-second-scene-extension-first-last-frame-control-and-4k-upscaling","status":"publish","type":"post","link":"https:\/\/youzum.net\/es\/google-ai-releases-gemini-omni-1-1-flash-40-second-scene-extension-first-last-frame-control-and-4k-upscaling\/","title":{"rendered":"Google AI Releases Gemini Omni 1.1 Flash: 40-Second Scene Extension, First\/Last Frame Control, and 4K Upscaling"},"content":{"rendered":"

Google has released Gemini Omni 1.1 Flash<\/strong> (gemini-omni-1.1-flash<\/code>), a production update to its native multimodal video generation and editing model. The release moves Omni from a capable generator to a directable one: scene extension now reads up to 10 seconds of prior context instead of a single final frame, first and last frames can be pinned to control camera movement, drafts render in 360p at a third of 720p cost, finals upscale to 4K, and video clips can be passed as references for character consistency. <\/p>\n

Gemini Omni Flash is built on three properties Google distinguishes from prior video models: native multimodality (text, image, audio, and video processed together), conversational editing through the Interactions API<\/a>, and world knowledge inherited from Gemini. Editing is stateful \u2014 you pass previous_interaction_id<\/code> and the model applies your change while preserving what you did not mention, without re-uploading the prior video.<\/p>\n

Is it deployable?<\/strong><\/h2>\n

It is available through the Gemini API<\/a> in Google AI Studio<\/a> and the Gemini Enterprise Agent Platform<\/a>, with Adobe, Figma Weave, GMI Cloud, and Runway already named as production users.<\/p>\n

\n<\/div>\n

\n

Scene Extension is the Main Change<\/strong><\/h2>\n<\/p>

Omni 1.1 analyzes up to 10 seconds of prior context<\/strong> when continuing a clip. Google states that previous models referenced only the final second. Extensions run in 10-second increments to a cumulative 40 seconds<\/strong>, and the model generates a 3\u201310 second continuation per call. Some final frames of the input are edited to make the seam continuous.<\/p>\n

The constraints are specific. Extension appends to the end of a clip only \u2014 no prepending, no mid-clip insertion. Uploaded input videos must be 10 seconds or shorter, unless you are extending a model-generated video in multi-turn. You cannot add new dialogue when extending an uploaded video where someone is speaking; spoken dialogue is supported in multi-turn extension via previous_interaction_id<\/code>.<\/p>\n

Keyframes and Video References<\/strong><\/h2>\n

You can now supply a first and last frame and have the model generate the continuous video between them, which is the mechanism behind orbits, dolly-zooms, and seamless loops. Prompts bind media to roles with tags: <FIRST_FRAME><\/code>, <LAST_FRAME><\/code>, <IMAGE_REF_N><\/code>, and <VIDEO_REF_N><\/code>.<\/p>\n

Video references accept a maximum of three clips, up to three seconds each<\/strong>, and work best for likenesses. Audio inside a video reference is ignored. Reasoning across multiple videos is not supported and may degrade output.<\/p>\n

Cost Control: Draft in 360p, Ship in 4K<\/strong><\/h2>\n

The resolution<\/code> parameter in response_format<\/code> takes 360p<\/code>, 720p<\/code> (default), 1080p<\/code>, and 4k<\/code>, with the top two upscaled. Google reports 360p previews generate up to 60% faster and at a third of the cost<\/strong> of 720p, based on system throughput of 360p versus 720p. That makes the draft-then-upscale loop the intended production pattern: iterate cheaply, render once.<\/p>\n

Pricing, provenance, and limits<\/strong><\/h2>\n

Input is $1.50 per 1M tokens<\/strong> (text, image, video, audio). Output is $9.00 per 1M text tokens<\/strong> and $17.50 per 1M video tokens<\/strong>. Video billing runs at 5,792 tokens per second of 720p, an effective ~$0.10 per second<\/strong> under standard pricing.<\/p>\n

Every generated video carries SynthID<\/strong> watermarking \u2014 invisible to viewers, programmatically detectable for provenance. Notable gaps: no system instructions, temperature, top_p<\/code>, stop sequences, or negative prompts (negatives go in the prompt text); voice editing is unsupported; audio references are unsupported; YouTube URLs cannot be used as a source. English is fully supported; other languages are unevaluated. For outputs above 4MB, use delivery=\"uri\"<\/code> and poll the Files API<\/a> until the file is ACTIVE<\/code>.<\/p>\n

Google names Adobe (Firefly), Figma Weave, GMI Cloud, and Runway as customers already running Omni Flash in production. The model is also live in Google Flow<\/a> for AI Plus, Pro, and Ultra subscribers, with scene extension in the Gemini app.<\/p>\n

Comparison<\/strong><\/h2>\n
\n<\/div>\n

\n

Key Takeaways<\/strong><\/h2>\n