ProductsSelected

Google launches Genie 4 on Live Interactions API

Why we picked it

Ephemeral WebRTC tokens remove a real security pain point for browser-based real-time AI apps, where embedding API keys client-side has been a common workaround. Developers building voice agents, live translation or interactive video products on Gemini should evaluate the Live Interactions API against alternatives like OpenAI's Realtime API. Feedback channels are open, so early adopters can shape which models and protocols get support next.

Google has launched Genie 4 as the first model available through its new Live Interactions API, which adds native WebRTC and WebSocket support to the Interactions API and Gemini API. WebRTC connections use ephemeral tokens, letting browsers establish direct real-time sessions without exposing long-lived API keys.

Image · Philipp Schmid
Body

The announcement, made by Philipp Schmid on X, explains that the Interactions API was designed so standard request-response calls, real-time sessions and long-running sessions all share one consistent interface. The team says it focused on getting these primitives right and plans to bring WebSocket and WebRTC protocols to more Gemini models over time.

The launch targets developers building real-time voice, video or agent applications that need low-latency bidirectional communication. Genie 4 can be tried now via Google AI Studio documentation, and Google is soliciting developer feedback on what is missing for agent and real-time app experiences.

Original · Philipp Schmid

Genie 4 is the first model we launched on the new Live Interactions API, bringing native WebRTC (with ephemeral tokens) and WebSocket support directly to Interactions and Gemini API. This is part of our ongoing work to improve Gemini API experience. When we started building the Interactions API, one of our main goals was making sure standard API calls and real-time and long running sessions share one consistent interface. We put a lot of effort into getting the primitives right, and we have big plans. I would love your feedback as we build this out: what do you want to see as we bring protocols like WebSockets and WebRTC to more Gemini models? What is still missing for building better agent experiences or real-time apps? In the meantime, you can try Genie 4 in the docs: aistudio.google.com/docs/genie

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Products" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include Genie 4, Gemini API, WebRTC, WebSocket.

Source attribution: curated from Philipp Schmid (@_philschmid).