AIToday
Large Language ModelsAI Safety & AlignmentLessWrong AIPublished: Sep 14, 2026, 13:00 JST

GPT-6-Astra hits 31% on 4-hop no-CoT tasks, others 1–3%

GPT-6-Astra hits 31% on 4-hop no-CoT tasks, others 1–3%

A Second Look Fellowship replication tested GPT-6-Astra on the same no-CoT items and protocol used for Fable 5, Opus 5, Opus 4.5, GPT-5.6-Sol, Gemini 3.1 Pro, Kimi k3, and Fable 5.1.

Ask the AI about this article →

Summaries like this, in your inbox every morning.

Get the latest Large Language Models news every morning

For example, today's edition would include:

  • Copado adds Headless mode to Agentia for Salesforce DevOpsSiliconANGLE AI · 12m ago
  • Jacob Coxon warns AI labs 'gambling with our lives'Fortune AI · 12m ago
  • Amodei's 'pacing the frontier' call splits software from chipsYahoo Finance AI · 3h ago

AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. Q&As are published on this page for other readers too.

Related Articles

Next articleBTL starts Guishan Plant 3 for AI server testing