Gemini Omni 1.2 Flash API: Now Available at 12% Off

Gemini Omni 1.2 Flash release date and API access

Google's November 5 release notes confirm general availability. Developers can use the exact ID gemini-omni-1.2-flash in the paid Gemini API and Google AI Studio. It is no longer an upcoming-model prediction.

This article first covered the October 28 community deployment report. The English original-post screenshot remains as a dated record of that earlier report:

Original community post. Current release status comes from Google's documentation.

Gemini Omni 1.2 vs 1.1: what changed?

The clearest upgrades are longer clips, more reference images and lower output cost. Google also reports improved character and subject consistency and doubled paid-tier rate limits; actual quotas depend on the account.

15 seconds is not the limit for every operation. Interpolation and each extension turn remain 3–10 seconds. The 30-second figure describes a video built through extension. 1080p and 4K are upscaled outputs.

Google says direct requests to gemini-omni-1.1-flash will automatically route to 1.2 at the new rates on December 3, 2026. This is Google's migration schedule; APIMaster route support and pricing remain subject to its marketplace card.

Official Gemini Omni examples and useful comparison prompts

The current Google guide now uses 1.2 in its examples. Its text-to-video prompt is: "A marble rolling fast on a chain reaction style track, continuous smooth shot." For local edits it recommends short instructions such as "Add a cat that jumps onto his lap, he begins to pet it. Keep everything else the same."

The earlier Google Omni launch demo includes "Make the sculpture out of bubbles." That is an Omni-family demonstration, not a controlled 1.2-versus-1.1 benchmark.

Three practical tests make the new capabilities visible:

• Character consistency: supply up to 6 authorized reference images of the same fictional character; ask for a continuous scene with a turn, a close-up and a wide shot. Check the face, clothes and props.
• Longer action: ask for a continuous shot of a marble hitting dominoes and landing in a bowl. Check whether the causal sequence survives to the end of a longer clip.
• Extension: continue a character walking toward a window. Check the join, motion direction, lighting and audio; measure total cost across all turns.

These are proposed acceptance tests, not claimed generation results or evidence of a measured quality win.

Gemini Omni 1.2 Flash API pricing

Google's standard paid rates are $1.50 / 1M input tokens, $9 / 1M text-output tokens including thinking, and $12 / 1M video-output tokens. The free tier does not include Omni 1.2.

At 720p, Google lists 5,792 video tokens per second:

5,792 × $12 / 1,000,000 = $0.069504 per second

Video output alone is about $0.69504 for 10 seconds or $1.04256 for 15 seconds, before input and text charges. The exact video-token reduction from $17.50 to $12 is 31.4%; Google's announcement rounds this to 30%.

APIMaster's 1.2 model card now has 1 public route at 12% off: input $1.32/M, text output $7.92/M and video output $10.56/M tokens. At 5,792 video tokens per second, the 720p video-output component is approximately $0.06116352/s, $0.6116352 for 10 seconds or $0.9174528 for 15 seconds; input and text cost extra. These are token-rate conversions, not a fixed all-inclusive per-second charge. The 1.1 card separately has 2 routes, with base prices from $0.055001/s. See the cheap API guide for resolution budgets.

How to use the released Gemini Omni 1.2 API

For official access, enable paid Gemini API billing and use the Google SDK. The following follows Google's current Interactions API example; it uses a Google API key, not an APIMaster key.

import base64
from google import genai

client = genai.Client()
interaction = client.interactions.create(
    model="gemini-omni-1.2-flash",
    input="A marble rolling fast on a chain reaction style track, continuous smooth shot.",
)
with open("marble.mp4", "wb") as output:
    output.write(base64.b64decode(interaction.output_video.data))

For APIMaster, register, create an API key, and open the 1.2 model card for 12% off. The current card lists POST https://apimaster.ai/v1/videos for gemini-omni-1.2-flash. Use its supported request profile; Google's SDK example above uses a different API contract. The older Omni 1.1 documentation remains specific to that version.

FAQ

Has Gemini Omni 1.2 Flash been released?

Yes. Google announced general availability on November 5, 2026, through the paid Gemini API. The exact model ID is gemini-omni-1.2-flash.

Can I buy Gemini Omni 1.2 Flash on APIMaster?

Yes. The November 7, 2026 check found 1 enabled public route at 12% off: input $1.32/M, text output $7.92/M and video output $10.56/M tokens. Open the 1.2 model card for access.

How long can Omni 1.2 videos be?

Generation and editing support 3–15 seconds. First/last-frame interpolation and each extension turn support 3–10 seconds; extension can build up to 30 seconds total.

What is the Gemini Omni 1.2 Flash price?

APIMaster charges $1.32/M input, $7.92/M text output and $10.56/M video output, 12% below Google's $1.50/$9/$12 reference rates. At 720p the video-output component converts to about $0.06116352/s, before input and text charges.

Is Omni 1.2 better than Omni 1.1?

It has confirmed longer generation/editing, up to 6 reference images, lower video-token pricing and higher paid-tier rate limits. Google reports improved subject consistency; this article does not claim an independent quality benchmark.

Sources and next step

APIMaster prices checked November 7, 2026 via the live marketplace API. Official release and specification sources retained from the November 6 review: Google release notes, model specification, developer guide, and official pricing.

Register → create an API key → https://apimaster.ai/v1. Select gemini-omni-1.2-flash on its model card for 12% off: $1.32/M input, $7.92/M text output and $10.56/M video output.