Google Gemini 4 Argon Tops OpenAI GPT-6.1 on Coding Benchmark

5 Oct 2026, 4:09 pm · 3h ago · 1 min read · OfficeChaiNew

Google’s newly released artificial intelligence model, Gemini 4 Argon equipped with Antigravity, has surpassed OpenAI’s flagship GPT-6.1 Sol on the Artificial Analysis Coding Agent Index for Codex evaluations. The benchmark ranking highlights Google's rapid technical progress in narrowing the competitive gap with OpenAI in automated software development and code generation benchmarks. The Artificial Analysis index tracks capabilities of autonomous coding agents across real-world repository tasks, debugging, and synthesis. This milestone reflects an escalating competitive race among frontier AI developers to dominate developer tooling, enterprise software workflows, and complex automated code completion ecosystems worldwide.