OpenAI's Astra for Law tops legal reasoning benchmark with efficiency gains
GPT-6.2 Astra configured for law achieves 20.8% full-credit pass rate on Harvey's LAB-AA, outpacing competitors while using less compute per task.
GPT-6.2 Astra configured for law achieves 20.8% full-credit pass rate on Harvey's LAB-AA, outpacing competitors while using less compute per task.
OpenAI releases GPT-6.2 Astra for complex work and GPT-6.2 Sol for everyday tasks, cutting input costs 20% while advancing on scientific and professional benchmarks.
AI assists in constructing a nonperturbative lattice formulation of particle physics' foundational theory.
AI vision models are collapsing simulation bottlenecks in robotics, enabling researchers to train robot behavior in a fraction of the time.
OpenAI and biotech firm Ligo launch a 100-task benchmark testing how AI agents interpret real experimental data, identify errors, and know when conclusions lack evidence.
Merge integrates GPT-6 Astra and Codex into its data-sync and AI-agent platform, achieving 99.9% tool-call reliability and automating customer relationship workflows.
GPT-6 Cyber achieves 48.5% on ExploitGym, up from 32.1% for GPT-5.6 Cyber, with improved alignment to stay within authorized scope during security testing.
OpenAI demonstrated Astra's capabilities in mechanical, electrical, and software design by building a functional device prototype with minimal human intervention.
Qodo integrates with OpenAI's Codex to give coding agents cross-repository context, automating P0 bug detection across distributed systems.
Terminal X's Capital Memory OS uses OpenAI's GPT-6.1 Sol to automate investment decision-making, matching senior investor judgment 88.6% of the time while cutting query costs by 30%.
GPT-6 Astra achieves 87.4% accuracy on multi-language database tasks, up from 26.8% for GPT-5.6 Sol, while COBOL migration accuracy reaches 97.9% at higher reasoning effort.