DeepMind reveals Gemini Robotics 2.5 and open safety benchmark "ASIMOV-jailbreak"

Model release Ars Technica AI Source language: English

The new model is up to twice as successful at some tasks.

Excerpt

The new model is up to twice as successful at some tasks.

Short excerpt from the source — open the original for the full text.

Sources

← Back to the daily brief