whatmodel.sh / video / Gemini Omni 1.2 Flash

Gemini Omni 1.2 Flash is a closed-weights video model from Google, released November 2026. Licence: proprietary.

Catalogue identity notes

Gemini API changelog, November 5, 2026: 'Gemini Omni 1.2 Flash generally available (GA): Released Gemini Omni 1.2 Flash (gemini-omni-1.2-flash)'. Stable model code gemini-omni-1.2-flash; up to 15-second video with native audio; 131,072 input tokens. Distinct from gemini-omni-1-1-flash (gemini-omni-1.1-flash), which Google will route to 1.2 from December 3, 2026, and from gemini-omni-flash (preview, now shut down). Corresponds to Arena 'Gemini Omni 1.2 Flash' on text_to_video, image_to_video and video_edit.

Model information sources

Release date ↗

80.9±1.7#1 of 94 video modelsfrom 3 benchmark results across 1 source

$ how it compares

The video models ranked either side of Gemini Omni 1.2 Flash, out of 141 with a score, then more from Google.

#modelscorewhereRAMreleased
1Gemini Omni 1.2 Flash Google81±2api $0.070/sec2026-11
2Gemini Omni 1.1 Flash Google80±2api $0.10/sec2026-08
3Wan 3.0 Prime Alibaba79±6api $0.14/sec2026-08
4Gemini Omni Flash Google79±2api $0.10/sec2026-05
5Wan 3.0 Alibaba78±1api $0.10/sec2026-08
more from Google
20Veo 3.1 Google68±1api $0.40/sec2025-10
22Veo 3.1 Fast Google67±1api $0.10/sec2025-10
28Veo 3 Fast Google65±2comparison only2025-07

see all video models · find the ones that fit your machine

$ what it's good at

text to video▆▄▃▆▆█▇▂▄·▄▅ #1 of 70 * · 55% of its score · 1 test
testresultvs the others in its rundifficultycountswhy
LMArena video1540.1 ✎▂▂▃▆█▃·▇▂▁█▃ beats 44/44avg 1154▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 55%typical
image to video▁▂·▂▅█▅▄▅▅▃▅ #3 of 51 * · 32% of its score · 1 test

Trails MiniMax H3, Gemini Omni 1.1 Flash here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
Arena image-to-video1503.0 ✎▁▁▁▁▄▇▃▇▂▃▅█ beats 42/42avg 1197▇▇▇▇▇▇▇▇▇▇▇▇▇ 32%typical
video editing▃··▅·▃···█▃▅ #3 of 10 * · 12% of its score · 1 test

Trails Dreamina Seedance 2.5, Wan 3.0 here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
Arena video edit1395.6 ✎▄··▂▂·▂··█▃▄ beats 9/11avg 1119▇▇▇▇▇ 12%typical

$ where to run it

Through an API

api · $0.070/sec · ai.google.dev ↗ · price sourceUSD per generated second of 720p video on the Gemini Developer API, paid Standard tier. Product: Gemini Developer API (Interactions API). API model: gemini-omni-1.2-flash (GA). Price basis: $12.00 per 1M video output tokens at Google's stated 5,792 tokens per second of 720p video (Google: 'approximately $0.07 per second'); 5,792 x $12.00/1M = $0.069504. Input ($1.50/1M text/image/video) and text output ($9.00/1M) are additional; 360p, 1080p and 4K use other token rates. Currency: USD. ai.google.dev/gemini-api/docs/* now redirects to aistudio.google.com/docs/* (pricing page fetched as https://aistudio.google.com/docs/pricing.md). Observed 2026-11-12. models.dev providers/google/models/gemini-omni-1.2-flash.toml (5056efa) has the same $12 video-output token rate but no per-second figure.

$ how the score was worked out · LMArena leaderboard dataset 99%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
LMArena leaderboard dataset3CC BY 4.0
prior · pull toward the middle
weighted average= 80.9

$ improve this page

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json