Google DeepMind's Gemini 4 Argon reportedly scored 97.5% on the ARC-AGI-3 benchmark, making Google the second lab after OpenAI to cross 95%. The result is highly setup-dependent, dropping to 49.8% in the standard harness. The model also trails rivals in math and software engineering benchmarks.
Google has launched Gemini 4 Argon, its long-awaited frontier AI model, after months of sitting out the frontier race. The company claims Argon offers frontier reasoning and better performance across complex workflows while reducing hedging, repetition, and cliches. Google says Argon outperforms GPT-6 Astra, Claude Fable 5.1, and Claude Opus 5.5 across benchmarks in areas including coding, cybersecurity, science, finance, and law. Google also says it is already using Argon internally.
Logan Kilpatrick, Senior Product Manager at Google DeepMind and Head of Google AI Studio, has issued a strong directive to the AI industry: every company building products on top of AI should establish its own proprietary benchmarks (standa<!--autointro-->
