Genie 4
GAReal-time interactive world generation on the Interactions API (v1beta): seed a world from a text prompt, a 16:9 reference image, or both, then steer the live 24 fps 16:9 video with synchronized 48 kHz stereo audio using keyboard controls (WASD/arrow keys) and mid-session text events, streamed over WebRTC or WebSockets. Google's changelog entry of November 19, 2026 is headed 'Genie 4 (genie-4) generally available (GA)'; the model page Versions block reads 'Stable: genie-4' and its 'Latest update' field says 'November 2026'; the deprecations page 'Genie models' table gives release date November 19, 2026, 'No shutdown date announced' and no replacement. Paid tier only (every free-tier cell reads 'Not available'). Single-column pricing with no dated schedule: input $1.50 per 1M (text / reference image); two output rates - text $9.00/1M and video $17.50/1M (audio included) - price_output_per_mtok holds the VIDEO rate, the same convention as gemini-omni-1-2-flash. Google bills emitted video output tokens at 240 tokens per frame (5,760 tokens per second at 24 fps), which it states is an effective ~$0.10 per second (~$6.05 per minute). No context caching price; Batch API, Flex and Priority inference are not supported. Session limits per the model page: output video 1280x720 (720p), up to 120 seconds per session, a single 16:9 reference image; serving regions United States, Western Europe and Japan. Google publishes no context-window or output-token limit and no knowledge cutoff, so context_window, max_output_tokens and knowledge_cutoff stay null. For offline fixed-duration clips Google points to gemini-omni-1-2-flash instead (guidance, not a replacement). Read first-party 2026-11-20 from aistudio.google.com/docs (changelog, model page, pricing, deprecations).
Genie 4 by Google costs $1.50 per 1M input tokens and $17.50 per 1M output tokens ($5.50/1M blended). It is generally available (GA).
Last verified: 21 Nov 2026 · sourced from official provider documentation
◎ You're on the watch list. We'll ping you the moment a model launches, changes price, or gets deprecated.
✕ Couldn't subscribe right now — please try again in a moment.
Free forever · no spam · unsubscribe anytime.
Source: Google official documentation ↗
Genie 4 — questions & answers
How much does Genie 4 cost?
Genie 4 is priced at $1.50 per 1M input tokens and $17.50 per 1M output tokens ($5.50/1M blended at a 3:1 input-to-output ratio).
Is Genie 4 deprecated?
No — Genie 4 is generally available (GA) and not currently scheduled for deprecation or retirement in our tracker.
Does Genie 4 support image or vision input?
Yes — Genie 4 accepts image input. Listed modalities: text, image, video-out, audio-out.
Can I self-host Genie 4?
No — Genie 4 is proprietary. Its weights are not publicly released, so it is available only through Google's API (or hosted partners), not for self-hosting.
What is the API model string for Genie 4?
The API model identifier for Genie 4 is "genie-4" when calling Google's API.
Track Genie 4 price & status changes
New models, price cuts, and deprecations — a short email when something actually changes. No spam, unsubscribe anytime.
The last few changes we caught — this is what lands in your inbox:
- RetiringxAI retired grok-build-0.1 - the early-access coding model is no longer served on the API
- New modelGenie 4 launched (GA) - real-time interactive world generation, $1.50 in / $17.50 per 1M video out, paid tier only
- New modelGemini 4 Neon launched (GA) - $3/$15 per MTok, paid tier only, 1M context
◎ You're on the watch list. We'll ping you the moment a model launches, changes price, or gets deprecated.
✕ Couldn't subscribe right now — please try again in a moment.
Free forever · powered by the same data on this page.