{"id":"gemini-omni-1-2-flash","slug":"gemini-omni-1-2-flash","name":"Gemini Omni 1.2 Flash","orgId":"google","family":"gemini-omni-1-2-flash","releaseDate":"2026-11-05","availability":"ga","openWeights":false,"license":"Proprietary","paramsTotal":null,"paramsSource":null,"contextWindow":131072,"knowledgeCutoff":null,"modelCardUrl":"https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/","hfRepo":null,"modalities":["text","image","video"],"aliases":["Gemini Omni 1.2 Flash","gemini-omni-1-2-flash","omni 1.2 flash","omni-1.2-flash","google/gemini-omni-1.2-flash","Google: Gemini Omni 1.2 Flash","gemini omni 1.2 flash"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-11-05","announced":"2026-11-05","blurb":"Google DeepMind's updated closed video generation model in the Gemini Omni Flash line. Generally available on the Gemini API and Google Cloud, with rollouts into AI Studio, Flow, Google Vids, Google Ads and the Gemini app. Versus Omni 1.1 Flash: native clips up to 15 seconds (from 10), multi-turn conversational edits, stronger reference adherence and scene consistency, draft-at-360p/720p then upscale to 1080p/4K, and a 30% price cut to $0.07 per second for 720p. Outputs carry an imperceptible SynthID watermark.","api":"live","openWeights":"no","facts":[{"label":"Role","value":"Closed video generation and editing model (text, image, short motion and audio references in; video out)"},{"label":"Clip length","value":"Native generation and editing up to 15 seconds; duration selectable in 1-second steps from 3 to 15 seconds (Omni 1.1 Flash was 10 seconds)"},{"label":"Editing","value":"Multi-turn conversational edits on the same clip (swap subject, background, lighting, aspect ratio) without regenerating from scratch"},{"label":"References","value":"Character, product or style images, short motion clips up to 3 seconds, and audio cues; holds faces, garments, packaging type and brand colors across angles"},{"label":"Resolution workflow","value":"Draft at 360p or 720p, then upscale selected clips to 1080p or 4K"},{"label":"Pricing","value":"$0.07 per second for 720p (30% below Omni 1.1 Flash, per Google); Provisioned Throughput available on Google Cloud"},{"label":"Availability","value":"Generally available on the Gemini API and Google Cloud; rolling out in Google AI Studio, Flow, Google Vids, Google Ads and the Gemini app (15-second clips in Gemini app for Google AI Ultra; Vids for Google accounts / Workspace)"},{"label":"Watermark","value":"Imperceptible SynthID on outputs"},{"label":"Weights","value":"Closed; none stated"}],"claims":["Lab claim: 30% lower price than Gemini Omni 1.1 Flash, at $0.07 per second for 720p video.","Lab claim: native 15-second video creation and multi-turn conversational editing, with improved reference following versus Omni 1.1 Flash.","Lab claim: stronger character, object and scene consistency across motion and camera transitions, with fewer unwanted jump cuts on continuous camera moves."],"sources":[{"kind":"lab","title":"Gemini Omni 1.2 Flash blog (Google)","url":"https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/","date":"2026-11-05"},{"kind":"lab","title":"Google on X: Gemini Omni 1.2 Flash","url":"https://x.com/Google/status/2118390406316728555","date":"2026-11-05"},{"kind":"lab","title":"Google AI on X: Gemini Omni 1.2 Flash","url":"https://x.com/GoogleAI/status/2118389961426874879","date":"2026-11-05"},{"kind":"lab","title":"GeminiApp on X: Omni 1.2 Flash rollout","url":"https://x.com/GeminiApp/status/2118406227533435391","date":"2026-11-05"}],"firstBenchmark":{"date":"2026-11-06","benchmarkId":"lmarena-t2v"}},"contextSource":"openrouter","listedSince":"2026-11-05","eligibleSince":"2026-11-06","organization":{"id":"google","name":"Google DeepMind","slug":"google","country":null,"url":"https://deepmind.google","color":"#4285F4","mark":"G"},"price":{"modelId":"gemini-omni-1-2-flash","provider":"google","inputPerM":null,"outputPerM":null,"blendedPerM":3.375,"tokensPerS":null,"observedAt":"2026-11-22","sourceId":"openrouter"},"answer":{"word":"Yes.","scope":"SOTA in Video generation (tied).","lead":"As of Nov 22, 2026, Gemini Omni 1.2 Flash is tied for state of the art in Video generation (since Nov 8, 2026; sources LMArena Text-to-Video; methodology v1.0). Gemini Omni 1.2 Flash is tied for SOTA in Video generation with Gemini Omni 1.1 Flash, within the margin of error, since 2026-11-08.","confidence":"medium"},"status":"BENCHMARKED","verdicts":[{"id":"v_overall_video-gen_gemini-omni-1-2-flash_2026-11-08_open_CO_SOTA","model":"gemini-omni-1-2-flash","model_name":"Gemini Omni 1.2 Flash","category":"video-gen","category_name":"Video generation","slice":"overall","state":"CO_SOTA","rank":1,"present_evidence":[],"missing_evidence":[],"coverage":null,"segments":[],"confirm":[],"margin":0,"confidence":"medium","valid_from":"2026-11-08","valid_to":null,"configuration":null,"rationale":"Gemini Omni 1.2 Flash is tied for SOTA in Video generation with Gemini Omni 1.1 Flash, within the margin of error, since 2026-11-08.","safeguard_flags":["two_signal_confirmed"],"co_sota_with":["gemini-omni-1-1-flash"],"challengers":[],"methodology":"1.0","label":"Co-SOTA · Video generation"},{"id":"v_pareto_video-gen_gemini-omni-1-2-flash_2026-11-06_open_SOTA","model":"gemini-omni-1-2-flash","model_name":"Gemini Omni 1.2 Flash","category":"video-gen","category_name":"Video generation","slice":"pareto","state":"SOTA","rank":1,"present_evidence":[],"missing_evidence":[],"coverage":null,"segments":[],"confirm":[],"margin":0,"confidence":"medium","valid_from":"2026-11-06","valid_to":null,"since_text":"since Nov 6, 2026","configuration":null,"rationale":"Gemini Omni 1.2 Flash leads the Video generation best-per-dollar frontier since Nov 6, 2026.","safeguard_flags":["pareto","two_signal_confirmed"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pareto · Video generation"}]}