Google DeepMindVideo modelResearch preview for trusted testers
Genie 4
Summary
Google DeepMind describes Genie 4 as an interactive world model that generates video, ambient sound, music, and dialogue together. The announcement is dated October 27, 2026 and says the research preview is for trusted testers. It does not document public API access, pricing, or numeric benchmarks.
The announcement identifies drift and physical, spatial, language, and temporal consistency as open research questions. The official model-card directory has no dedicated Genie 4 card.
On this page
All specifications
Input and output
- Capabilities
- Real-time interactive world simulation with video, ambient sound, music, dialogue, and character motion
- Source: Capabilities
- Modality details
- Interactive video with generated music, ambient sound, and dialogue
- Source: Modality details
Licence and access
Further details
- Model card
- The official DeepMind model-card directory does not list a dedicated Genie 4 model card.
- Source: Model card
- Stated limitations
- Google identifies ongoing work on drift and on physical, spatial, language, and temporal consistency.
- Source: Stated limitations
Sources and data date
| Type | Evidence and data date |
|---|---|
| Research status | Research date October 29, 2026. 58 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Not found in the inspected sources | Genie 4 model card Note The official model-card directory has no dedicated Genie 4 card. |
| Not published | Numeric benchmark results Note The official announcement gives no numeric Genie 4 benchmark results. No Genie 3 or other model scores are attributed to Genie 4. |
| Not published | Public API, pricing, and general availability Note The reviewed source describes a trusted-tester research preview and future availability plans, not public API access or current pricing. |
| Not found in the inspected sources | Vals rows in captured payloads Note Nine complete Vals payloads dated 28 October were checked; no exact model ID or unambiguous model label. Limited to these saved benchmarks. |
| Unresolved | Uncaptured Arena Code/Speech leaderboards Note Text, Agent including Code/Chat/Work, and twelve WebDev views were captured. A separate general Code and Arena Speech leaderboard were not saved as complete tables. |
| Unresolved | Artificial Analysis catalog reconciliation Note The full current AA model and benchmark catalog was not re-reconciled for this profile in this sidecar. This sidecar makes no AA absence claim; native AA imports remain authoritative. |
| Additional source | Google DeepMind Genie 4 announcement (retrieved October 29, 2026) · Google DeepMind Genie 4 announcement · Editorial description reviewed October 29, 2026 |
| Additional source | Google DeepMind model-card directory (retrieved October 29, 2026) · Google DeepMind model-card directory · Editorial description reviewed October 29, 2026 |
| Additional source | Model metadata (retrieved October 29, 2026)12 |