Gemini Omni 1.2 Flash (Video Edit)

Gemini Omni 1.2 Flash edits a 3-15 s video from a plain instruction: swap objects, backgrounds, lighting or style, and keep the rest.

Google Gemini Omni 1.2 Flash edits a video from a plain instruction: swap a product or the background, change the lighting, weather, style or wardrobe, add or remove an object, rewrite on-screen text. The result is as long as the input (3-15 seconds) and keeps what the prompt does not mention; short instructions work best, e.g. "Make the car red. Keep everything else the same."

Inputs: prompt (required), video (required; MP4 or MOV of 3-15 seconds, up to 50 MB), reference_images (optional, up to 2, e.g. a product to put in), resolution (720p by default) and aspect_ratio.

It does not make a video longer (use flymyai/gemini-omni-1_2-flash-extend) and does not change voices. Google refuses videos of real public figures.

Price: per second of video by resolution, $0.024 at 360p, $0.07 at 720p, $0.105 at 1080p and $0.209 at 4K, plus the input and thinking tokens Google reports ($1.50 and $9.00 per 1M tokens): each second of 720p input video is 5,792 input tokens, about $0.009. An 8-second edit at 720p costs about $0.63. A refused or failed run costs nothing.

Price: From $0.024 per second of video, depending on resolution; an edit is as long as its video (3-15 s); input tokens $1.50 per 1M, text and thinking tokens $9.00 per 1M.

Run Gemini Omni 1.2 Flash (Video Edit) via API

curl -X POST https://api.flymy.ai/api/v1/flymyai/gemini-omni-1_2-flash-edit/predict \
  -H "X-API-KEY: $FLYMYAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Make it night with neon reflections. Keep everything else the same.","video":"https://storage.googleapis.com/open-inference-results/agent-files/1585/40ecb0fd73ed4b5eb3b34b53b3b7e290/bytedance-seedance-2_0-fast-text-to-video_a64b.mp4","resolution":"720p"}'

API documentation · Create a free account · Pricing

Related models