Live updates
Google opens Genie 4 to developers in the Gemini API at about $6 a minute
First seen on X 36 hours ago@OfficialLoganK ♥ 2,04523 views
Google has opened Genie 4, the world model it rolled out to AI Ultra subscribers a day earlier, to developers through the Gemini API and Google AI Studio. From a text prompt or a reference image, apps can generate a continuous 720p, 24fps world that users walk through with the keyboard while text prompts change the weather, lighting or events on the fly.
Genie 4 requires a paid-tier API key and does not support the free tier or the Batch API, according to Google's Nov 19 blog post and Gemini API docs. Video output is billed at 240 tokens per frame and $17.50 per million tokens, or roughly $0.10 a second and $6.05 a minute, with audio included; that is about $363 for an hour. Text and image input costs $1.50 per million tokens. Closing a browser tab does not immediately stop generation on the server, Google warns: "Always call complete() to end the session and stop billing." The model is served only from U.S. data centers, and Google says round-trip latency above 150 ms noticeably delays keyboard steering. Logan Kilpatrick, product lead for Google AI Studio, called it "our first world model for developers to build with." Every output carries a SynthID watermark.
Comments 0
Select a sentence in the story (long-press on phones) to comment on it.
“Introducing Genie 4 in the Gemini API + @GoogleAIStudio , our first world model for developers to build with and create new world model experiences 🔮 Try it out:”
