TechnologyGrok 4.7 reportedly still fails most of the time despite training on multi-hour tasks
The New Stack reports that SpaceXAI trained Grok 4.7 on multi-hour tasks to improve agent stamina.
TLDR
The New Stack says Grok 4.7 was built to work for hours but still fails most of the time. SpaceXAI trained the model on multi-hour tasks to improve agent stamina, the outlet reports.
Grok 4.7 reportedly still fails most of the time despite training on multi-hour tasks
The New Stack reports that SpaceXAI trained Grok 4.7 on multi-hour tasks to improve agent stamina.
TLDR
The New Stack says Grok 4.7 was built to work for hours but still fails most of the time. SpaceXAI trained the model on multi-hour tasks to improve agent stamina, the outlet reports.