Tweet: LLM post-training used to mean fine-tuning to a downstream task Robotics has been stuck in this setting, needing task-specific fine-tuning for best performance π07 changes this: It works out of the box & outperforms fine-tuned specialists Details: http://pi.website/pi07 @chelseabfinn: "LLM post-training used to mean fine-tuning to a downstream task Robotics has been stuck in this setting, needing task-specific fine-tuning for best performance π07 changes this: It works out of the box & outperforms fine-tuned specialists Details: http://pi.website/pi07 https://x.com/chelseabfinn/status/2044855690414694567/video/1" Dual-arm robotic hardware with black and silver components labeled 006-WR0 and G1-01-1400 uses one gripper to hold a dark green zucchini steady while the other wields a chef's knife with green tape on the handle to produce repeated downward cuts on a blue textured cutting board. Round zucchini slices pile up across the board surface as fixed on-screen text reads "autonomous, 5x speed" in the bottom-left corner and a white π logo stays visible in the bottom-right corner.