Qianjue launches Polibrain, a world model that predicts task-relevant state changes instead of video frames
Robot-brain developer Qianjue launched Polibrain, a world model that predicts task-relevant state changes instead of future video frames. It is adapted to quadruped, humanoid and wheeled robots, according to Pandaily.
Qianjue, a robot-brain developer, has launched Polibrain, a world model that predicts task-relevant state changes instead of future video frames. Pandaily reported the launch.
What was announced
- Model: Polibrain, described as a world model.
- Developer: Qianjue, which builds robot brains.
- Prediction target: task-relevant state changes, not future video frames.
- Embodiments: adapted to quadruped, humanoid and wheeled robots.
How it differs from frame-prediction approaches
Many world models forecast what a scene will look like next by generating future video frames. Polibrain instead predicts the state changes that matter for the task at hand, according to the report.
The design is also presented as cross-embodiment. Polibrain is adapted to three robot body types: quadruped, humanoid and wheeled.
What this means
Polibrain's stated design choice is to predict the state changes that bear on a task rather than reconstruct future video frames. Its adaptation to quadruped, humanoid and wheeled robots shows Qianjue aiming at more than one robot form with a single world model. The launch is notable for that framing, which pairs task-focused prediction with support for several body types.
Related Articles
Google's Gemini 4 Argon targets full robotics pipeline: 87% on ERQA, but third on AA-Robotics at 55/100
Google DeepMind is positioning Gemini 4 Argon as one model that controls robots, builds simulated training environments, and grades training data. Google claims 87% on the ERQA benchmark, ahead of Claude Opus 5.5 at 82.3%. Artificial Analysis's AA-Robotics test, published a day later, placed Argon third at 55 out of 100.
Google DeepMind: Gemini 4 Argon scores 87% on ERQA embodied reasoning, targets full robotics stack
Google DeepMind says Gemini 4 Argon scores 87% on the Embodied Reasoning (ERQA) benchmark using pure reasoning, ahead of every other frontier model. A new robotics post demonstrates robot control, Real2Sim reconstruction, physical understanding, and self-improvement workflows. Pricing, context window, and availability were not disclosed.
NVIDIA expands Cosmos with world model training tools, simulation checkpoints and Data Factory blueprints
NVIDIA has expanded the Cosmos platform with new training tools, simulation checkpoints and Cosmos Data Factory blueprints. The company says Cosmos ranks in the top five on six different open model leaderboards.
Ollama v0.40.9 adds Phonon speech generation on Apple Silicon and d3 decision models
Ollama v0.40.9 can generate speech with Phonon on Apple Silicon, with audio streaming through an OpenAI-compatible /v1/audio/speech endpoint. The release also adds the d3 family of decision models in 2B, 4B, 9B, and 27B sizes and brings Flint to Linux and Windows.
Comments
Loading...