Google Gemini Integration

Google's most capable multimodal AI

Gemini is Google's family of multimodal models that natively understand text, images, audio, and more. Built for both speed and depth, Gemini brings Google-scale AI to everyday workflows.

Kohala uses Gemini for image understanding and as a resilient fallback for illustration generation, and its Veo model powers cinematic AI video generation — so agent tiles, koans, reports, and video clips always have a visual.