Google Releases Gemini Omni 1.2 Flash with 50% Longer Clips and 30% Lower Cost

By ▲ Marktechpost AI (@Marktechpost)Original title: ▲ Google just released Gemini Omni 1.2 Flash, making AI video clips 50% longer and 30% cheaper.
78AI Score

Google launched Gemini Omni 1.2 Flash, raising native clip length from 10 to 15 seconds at $0.07 per second of 720p, with multi-turn editing across full clips. The workflow is draft-edit-upscale: generate at low resolution, refine with follow-up prompts, then upscale to 1080p or 4K. Google published no benchmark scores and admits fast sports action and complex physics remain weak.

Full Report

▲ Marktechpost AI · 2026-11-05 23

▲ Google just released Gemini Omni 1.2 Flash, making AI video clips 50% longer and 30% cheaper.

The key idea is a draft, edit, upscale loop. You draft at 360p or 720p, fix details with follow-up prompts, then upscale the best take to 1080p or 4K. Multi-turn editing now works across full 15-second clips, so you change 1 thing instead of re-generating everything.

The headline numbers: native clips up to 15 seconds (up from 10 on Omni 1.1), length set in 1-second steps from 3 to 15, and $0.07 per second of 720p video. Veo 3.1 tops out at 8 seconds per clip. It is a trade-off though: Google has published 0 benchmark scores for 1.2, and says fast sports action and complex physics still need work.

Generally available now on the Gemini API and Google Cloud, closed weights, about $1.05 for a full 15-second 720p clip......

Full analysis:

推文配图