Claude Opus 4.6 has emerged as the new top-ranked AI model, holding the number one position until Artificial Analysis completes its benchmarking of OpenAI's Codex 5.3, which is anticipated to lead in coding tasks. Anthropic's latest model leads the Artificial Analysis Index, a comprehensive suite of ten tests covering coding, agent-based tasks, and scientific reasoning. It achieved first-place rankings specifically in agent-based workflows, terminal coding, and physics research problems.

The cost to run the full test suite was $2,486, exceeding the $2,304 required for GPT-5.2 at its maximum reasoning capacity. Opus 4.6 consumed approximately 58 million output tokens, which is double the amount used by Opus 4.5 but significantly lower than GPT-5.2's 130 million. The higher overall cost is attributed to Anthropic's token pricing, set at $5 per million input tokens and $25 per million output tokens.

Opus 4.6 is accessible via the Claude.ai application, as well as through Anthropic's API, Google Vertex, AWS Bedrock, and Microsoft Azure.