Google Deepmind Genie Puts
Genie 4 confirms world models are now DeepMind's second front alongside Gemini, and the hybrid architecture is the interesting part: a planner running in the background to fight the drift that plagues pure generative video. The Decoder's framing is apt — this is model-based agency, an idea older than LLMs, finally made viable by scale.… more
In our view the honest caveat is the one stated: visual degradation in longer sessions means this is nowhere near replacing game engines, and calling it early research is fair rather than hedging. The bigger prize isn't interactive entertainment anyway — it's cheap, controllable training environments for agents and robotics. The real milestone to watch is the first stable hour-long session; whoever ships that changes the conversation.
Google Deepmind's Genie 4 puts a modern spin on agent designs that predate LLMs
Google Deepmind has introduced Genie 4, a real-time world model that simulates interactive 3D worlds with sound and talking characters. Gemini models help plan in the background to keep the simulation consistent. Genie 4 is still an early research model, and longer ses…
Source:
The Decoder
Discussion (0)