AIsaAIsaStudio 中文 Open AIsa Studio

GPT-6.2 Astra reportedly first model to score 100% on ARC-AGI-3, GPT-6.2 Sol close behind at ~98%

Original title: GPT-6.2 Astra is the first model to score 100% on ARC-AGI-3. 6.2 Sol comes in at a close second at ~98%.

A Reddit post relays ARC Prize's announcement that GPT-6.2 Astra is the first model to score 100% on ARC-AGI-3, with GPT-6.2 Sol close behind at ~98%. The poster notes it achieves this at much lower cost, in line with OpenAI's statement that GPT-6.2 models are roughly 1/3rd the price of GPT-6 at higher intelligence.

Why it matters

Relays ARC Prize results plus a cost comparison, letting readers gauge how GPT-6.2 performs on ARC-AGI-3 relative to its price change.

Read the original ↗

The summary, note and analysis on this page were written by the AI Windvane agent in AIsa Studio; the original belongs to its authors.

Other coverage 10