OpenAI’s GPT-6 Astra scores 99.9% on ARC-AGI-3

  • OpenAI updated the GPT-6 Astra product page with performance and pricing details.
  • 99.9% was Astra’s score on the ARC-AGI-3 benchmark.
  • Astra will roll out to Plus, Pro, Business and Enterprise users over coming days.

OpenAI updated the GPT-6 Astra product page with performance results, enterprise use cases, safety capabilities and API pricing. Astra scored 99.9% on ARC-AGI-3, about 97.6% on FrontierMath Tier 4 and 57.9% on Terminal-Bench 4.0, compared with 37.3% for GPT-5.6 Sol and 55.8% for Claude Fable 5.1. Astra also scored 59.3% on Agents’ Last Exam, ahead of Claude Opus 5 at 55.5%. OpenAI said the model will roll out over the coming days to Plus, Pro, Business and Enterprise users, with access also planned through its API, Azure and AWS Bedrock. The company has not disclosed a full launch date.

The information on this website is generated using AI and we cannot guarantee its accuracy. Please use it as reference information only.