Science and breakthroughs
Eval
Current state
The desk has not filed a dated note for this topic.
OpenAI’s BioBasicsBench puts GPT-6 Astra at 66.7% on wet-lab tasks
OpenAI and Ligo’s 100-task BioBasicsBench scores GPT-6 Astra at 66.7% mean reward at xhigh, versus 43.8% for GPT-5.6 Sol.
A topic is a tag already on a story. It is listed here when at least two stories use it.
Science and breakthroughs
METR says Anthropic is improving well-specified tasks at about 2× prior pace
Questions
What is on this page?
This page lists 3 stories tagged eval.
What is the current state?
The desk has not filed a dated note for this topic.
Where are the sources?
Each story links the sources it used.