skills.rest
amoai-tech
amoai-techCOMMUNITY
Explore all skills by this author →
GitHub

gemini-omni-flash-api

Generate, edit, and extend videos using Gemini Omni 1.1 Flash via the google-genai SDK.

1|Updated Sep 15, 2026
Search Tags:#gemini#ffmpeg#video generation#video editing#text-to-video#google-genai#video extension
One-click install
npx skills add https://github.com/amoai-tech/mdeai --skill gemini-omni-flash-api-amoai-tech
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: gemini-omni-flash-api
Source: https://github.com/amoai-tech/mdeai/tree/main/.claude/skills/gemini/references/official/gemini-omni-flash-api
Command: npx skills add https://github.com/amoai-tech/mdeai --skill gemini-omni-flash-api-amoai-tech

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, ffmpeg, ffprobe, and includes scripts (resource) components.

What problem does it solve? Creating and editing videos with generative AI requires handling media uploads, prompt structuring, resolution and duration configuration, and multi-turn state management, which is complex to wire up manually against the Gemini API. ## Core Features & Use Cases - Video Generation: Text-to-video, first-frame-to-video, first-and-last-frame transitions, and looping videos at resolutions from 360p up to 4K. - Video Editing & Extension: Edit existing videos (up to 10s), extend videos by 10s per turn up to 40s total, and regenerate audio from scratch by stripping the source audio track. - Media Pre-processing: Inspect videos with ffprobe and trim, scale, or normalize large source files with ffmpeg before upload to avoid slow uploads and OOM errors. - Use Case: A content creator uploads a product photo as the first frame, provides a style reference image, and generates a 10-second 1080p promotional clip, then extends it with a new scene using the previous interaction ID. ## Quick Start Generate a 10-second 1080p video of a sunset over the ocean and save it to media/sunset.mp4.

Frequently Asked Questions about gemini-omni-flash-api

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a video from a text prompt with Gemini Omni Flash?▼

Run scripts/video/generate_video.py with your prompt and an output path, for example: ./scripts/video/generate_video.py "A cat drinking tea" --output media/cat.mp4. Set the GEMINI_API_KEY environment variable first, and optionally pass --resolution, --aspect-ratio, and --duration.

How do I extend a video with the Gemini API?▼

Pass an existing video with --extend or reference a prior generation with --previous-interaction-id. Each turn extends by up to 10 seconds, up to a total of 40 seconds, and you can include reference images or videos when omitting the explicit task parameter.

What resolutions does Gemini Omni 1.1 Flash support?▼

It supports 360p, 720p, 1080p, and 4k output in both 16:9 landscape and 9:16 portrait aspect ratios. The default is 720p, and 4K requests take longer so a higher --timeout value is recommended.

Why does my video edit return empty output with zero tokens?▼

Uploading videos for edits or extensions is not available in the EEA, Switzerland, the United Kingdom, and some US states. If a video-to-video request completes quickly with total_output_tokens of 0, this regional restriction is the likely cause.

How do I make Gemini Omni Flash regenerate audio instead of keeping the original?▼

Strip the audio stream from the source video before upload using --strip-audio with prep_video.py or generate_video.py. If any audio stream remains, the model preserves or modifies it instead of generating new audio from scratch.

What are the limitations of video extension in Gemini Omni Flash?▼

Input videos for extension must be 10 seconds or less, and you cannot add new spoken dialogue when extending an uploaded video where someone is talking. Dialogue generation is only supported when extending previously generated videos via multi-turn interaction IDs.

Related Skills

Explore More Design & Creative Skills→
TeamWiseFlowTeamWiseFlow
8.5k

video-producer

Produces videos end-to-end from script to render, or edits existing clips with ffmpeg-based tools.

Official
Advanced
artokunartokun
530

director

Orchestrate a full short film production pipeline from a text story into a final video.

Community
Advanced
lvzhengbinlvzhengbin
2

video-script-storyboard

Generate video production packages with scripts, storyboards, and AI prompts.

Community
Advanced
hasnahasna
24

video-generation

Generate videos from text prompts using Google Veo 3.1, OpenAI Sora, and Runway.

Official
Intermediate