Gemini Omni Flash generated immediate interest after Google I/O, but the key limitation is simple: the Omni Flash API is not generally available for developers yet.
Gemini Omni separates the experience into three clear tools: prompt optimization, AI image generation, and AI video generation.
What is available now
- Gemini Flash can rewrite rough ideas into high-quality English generation prompts.
- AI Image Generator supports Text to Image and Image to Image.
- Gemini Omni Video Generator supports Text to Video, Image to Video, and Reference to Video.
What is not available yet
The missing piece is official Omni Flash API access with native multimodal conversation over video outputs. Until that changes, product builders should avoid promising exact Omni Flash behavior.
Recommended workflow
Start with the user idea, optimize it into English, create or edit an image when visual control matters, then generate video from text, image, or reference video.
