OpenAI says GPT-6 Astra is state-of-the-art on FrontierMath Tier 4, ARC-AGI 3, and TerminalBench-4.0, and also leads on Terminal-Bench Science 0.1 and HealthBench Pro—highlighting agentic terminal and scientific discovery strength.
Perplexity says GPT-6 Astra scored 0.682 on WANDR at about $11.98 per task—the highest of any model they tested. Versus Claude Fable 5.1 it is ~13.5% higher score at ~6.1% lower cost; versus Opus 5 it is ~27.0% higher score at only ~3.3% higher cost.
ARC Prize measured OpenAI GPT-6 Astra on ARC-AGI-3 Semi-Private: about 62.7% for ~$26k with the Standard harness and 99.9% for ~$19k with a Provider Adapter harness that preserves opaque reasoning state. Astra beat the human action-efficiency baseline on 96% of levels and built custom algebraic notes and tools.
OpenAI says GPT-6 Astra is state-of-the-art on Agents’ Last Exam, AutomationBench, and ScreenSpot Pro—benchmarks for cross-profession computer workflows and screen grounding—adding computer-use evidence beyond the already announced TerminalBench / ARC-AGI / FrontierMath results.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.