ARC-AGI-3

Google DeepMind's Gemini 4 Argon scored 97.5% on ARC-AGI-3 at a cost of $4,800 using the Provider Adapter Harness, according to ARC Prize. Under the Standard Harness it falls to 49.8%, and Epoch AI places it eighth on its Math ECI and sixth on its SWE ECI.

GPT-6 Astra completed Pokémon FireRed in 18 hours 12 minutes, five times faster than its predecessor, and scored 62.7% on ARC-AGI-3 versus 7.78% for GPT-5.6 Sol. The model also ran a 141-hour Minecraft session and finished Fallout 3 in roughly 59 hours, according to independent testers.

A Latent Space report describes GPT-6 Astra, a new OpenAI model the blog says can autonomously handle AI engineering tasks—training models, labeling data, deploying systems—at an estimated cost of under $6 per hour. The claims, including 97.6% on FrontierMath and 99.9% on ARC-AGI-3, come from independent blog testing rather than an official OpenAI announcement.

Nvidia researchers report that a custom agent harness with a supervisory component drove Claude Opus 5 from a 30% baseline score to a perfect 100% on the ARC-AGI-3 benchmark. The findings add to mounting evidence that scaffolding, not the underlying model, is the dominant factor in long-horizon agentic performance.

OpenAI says GPT-5.6 Sol scores 38.3 percent on ARC-AGI-3 using a custom Responses API configuration, edging out Claude Opus 5's 30.2 percent. Under the official benchmark harness, GPT-5.6 Sol's score drops to 7.8 percent, raising questions about what the comparison actually measures.

Claude Opus 5 scored 30.2 percent on the ARC-AGI-3 benchmark, nearly four times the previous record of 7.8 percent set by OpenAI's GPT-5.6 Sol (Max). The ARC Prize team attributes the leap to genuinely stronger reasoning, though an independent test on a separate puzzle benchmark showed far smaller improvements.