Gemini Omni 1.1 Flash Lets You Build with More Control

Quick Answer

It's official: Google announced Gemini Omni 1.1 Flash on August 27, 2026. The official blog post is titled "Gemini Omni 1.1 Flash lets you build with more control" — introducing "a new suite of creative controls and generative video capabilities" that make Omni 1.1 "production-ready for professional use via the Gemini API in Google AI Studio." The Gemini API docs now list a new model ID — gemini-omni-1.1-flash, with no -preview suffix. Google's pricing page marks it Generally Available. Headline upgrades, in Google's own words: extend a scene, first and last frame interpolation, crisp 4K upscaling, and faster prototyping (360p drafts). Google's own video generation overview now presents Omni as the model that replaces Veo 3.1 in the Gemini app.

One naming note: Google's official ordering is "Gemini Omni 1.1 Flash" (matching its Gemini 2.5 Flash convention), while most of the community — and this site — had been saying "Omni Flash 1.1." Same model.

Key Takeaways

Announced August 27, 2026 via Google's developer blog and @Google on X.
It graduated to GA. The new model string is gemini-omni-1.1-flash — no -preview. After five months of preview status, this is the production green light.
The 10-second wall cracks. Base clips are still up to 10 seconds, but scene extension now continues a video using 10 seconds of prior context (versus Veo's 1 second), up to 40 seconds total.
720p pricing holds at ~$0.10/second. Billing is token-based ($17.50 per 1M video output tokens at 5,792 tokens/second of 720p). The new tier is 360p drafting: up to 60% faster at one-third the cost — iterate cheap, then upscale.
1080p and 4K arrive via upscaling, not native generation. Default output remains 720p.
The old gemini-omni-flash-preview still works and is still listed at the same price — but it's now the legacy string. Plan your migration.

What's new in 1.1

Everything below is from Google's official announcement and the updated API docs:

Scene extension (up to 40s). Extend a finished clip in 3–10 second continuations, with the model analyzing the last 10 seconds for continuity. This is the feature Google is pushing to consumers too — it's rolling out to AI Plus, Pro, and Ultra subscribers in the Gemini app.
First/last frame control. Specify the starting and ending frames of a shot and the model interpolates between them — the storyboard-style control Veo users had been asking for in Omni.
360p draft mode. Generate drafts up to 60% faster at one-third the cost, then re-render or upscale the keeper. This changes the iteration economics more than any price cut would have.
4K upscaling. Output can be upscaled to 1080p or 4K for delivery.
Video references. Drop in up to 3 seconds of reference video to carry movement, visual context, and character consistency into a new generation.
Multi-turn and video-to-video editing. Conversational editing is now stateful across turns, and you can upload an existing video (≤10s) and edit it — both marked "New" on Google's overview page, alongside an avatar feature in the Gemini app.

Aspect ratios remain 16:9 (default) and 9:16.

Pricing

Per Google's pricing page, gemini-omni-1.1-flash is Generally Available and billed by output tokens: $17.50 per 1M video output tokens, at 5,792 tokens per second of 720p video — effectively ~$0.10 per second at 720p, unchanged from the 1.0 preview. The full resolution ladder: $0.03/second at 360p, $0.10 at 720p, $0.15 at 1080p, and $0.30 at 4K. Input runs $1.50 per 1M tokens across text, image, video, and audio.

So the leaked 1.1 pricing table we tracked before launch turned out to mean new tiers, not a price cut — the headline rate held, and the news is the cheap draft lane.

How the pre-launch leaks scored

We tracked this release before it was announced, so here's the honest scorecard on the five questions we said would matter:

Does it go GA? ✅ Yes — gemini-omni-1.1-flash, production-ready, no preview suffix. The Pre-GA listing read correctly.
Does the price move? ➖ Headline 720p stays at ~$0.10/second; the leaked new pricing table was for the added 360p draft tier.
Does 720p become 1080p? ➖ Partially — 1080p and 4K ship as upscaling, native generation stays 720p.
Does the 10-second cap lift? ✅ Yes, via scene extension to 40 seconds total.
Does avatar mode ship? ✅ Yes — an avatar feature is live-flagged "New" on Google's overview page for the Gemini app.

The "omni-1p2" checkpoint leak was also consistent with what shipped: a 1.1 product version riding newer foundation weights.

Migrating from the 1.0 preview

If you built on gemini-omni-flash-preview, nothing breaks today — the preview string is still listed and still priced identically. But it's now the deprecated path in all but name:

Swap the model string to gemini-omni-1.1-flash. If your calls are wrapped behind your own interface — the advice from our API status post — this is a config change.
Re-test durations. Extension requests behave differently (3–10s continuations against 10s of context), and input videos for editing must be ≤10 seconds.
Adopt the draft loop. Prototype at 360p, then final-render at 720p and upscale. For iterative prompt work this cuts spend by more than half.
Full parameters and runnable code are in our API tutorial, which we're updating for 1.1.

If you'd rather not manage model-version churn at all, an independent platform like Omni Flash absorbs it for you — prompt, generate, and refine in the browser, with no model strings or migration windows to manage.

FAQ

Is Gemini Omni 1.1 Flash released?

Yes. Google announced it on August 27, 2026, under the official name Gemini Omni 1.1 Flash. The model ID gemini-omni-1.1-flash is Generally Available in the Gemini API and Google AI Studio, and the model is rolling out in Flow, the Gemini Enterprise Agent Platform, and the Gemini app.

Is "Omni Flash 1.1" the same as "Gemini Omni 1.1 Flash"?

Yes — same model. Google's official naming puts the version number in the middle ("Gemini Omni 1.1 Flash"), following its Gemini 2.5 Flash convention; "Omni Flash 1.1" is the ordering most of the community used while tracking the release pre-announcement.

What's new in Omni 1.1 Flash vs 1.0?

Scene extension up to 40 seconds (with 10 seconds of context), first/last frame control, a 360p draft mode at roughly one-third the cost, 4K upscaling, 3-second video references, stateful multi-turn editing, and video-to-video editing of uploaded clips. Native resolution remains 720p; 1080p/4K come from upscaling.

Did the price change?

The headline rate didn't: 720p output still works out to about $0.10 per second ($17.50 per 1M video output tokens at 5,792 tokens/second). What's new is the resolution ladder — $0.03/second for 360p drafts (up to 60% faster), $0.15 for 1080p, and $0.30 for 4K — which changes iteration economics more than any headline cut would have.

Try the Omni Flash studio → — an Omni-era generation and conversational-editing workflow in the browser, no API key required.
Read: How to extend Gemini Omni 1.1 Flash videos to 40 seconds → — prompts, API code, cost, and the 40-second ceiling.
Read: First and last frame in Gemini Omni 1.1 Flash → — orbits, zooms, and loops from two stills.
Read: Draft at 360p, upscale to 4K → — the 1.1 cost ladder and the cheap-then-sharp loop.
Read: Gemini Omni Flash API tutorial → — model ID, pricing, parameters, and runnable code.
Read: Gemini Omni vs Veo 3.1 → — what the app-level replacement means for each model.
Read: Gemini Omni Flash alternatives → — where 1.1 stands against Seedance 2.0 and Kling 3.0.
Compare credit packs → — pick a plan and start creating today.