GPT-6 Astra full benchmarks: ARC-AGI-3 jumps from 7.8% to 98.6%, GPQA 96%

kimmonismus · x · 2026-09-04

kimmonismus shared the full benchmark suite for GPT-6 Astra: FrontierMath Tier 4 v2 at 97.6%, DeepSWE v1.1 at 74.1%, BenchCAD 95.9%, GPQA Diamond 96%, ExploitBench 100%, and ARC-AGI-3 leaping from 7.8% to 98.6%. OpenAI plans to bring the model to paying ChatGPT subscribers and its API over the next several days.

Related event: GPT-6 Astra Launches with Benchmark Leaks: ARC-AGI-3 Hits 98.6%(41 posts)→

Original post →

More from Models

Models channel →