AnnouncementsFeb 5, 2026Introducing Claude Opus 4.6We’re upgrading our smartest model. Across agentic coding, computer use, tool use, search, and finance, Opus 4.6 is an industry-leading model, often by wide margin.

Anthropic announced the release of Claude Opus 4.6, the latest update to its most advanced large language model, on February 5, 2026. The new model introduces significant improvements in agentic coding, multi-step reasoning, and, for the first time in its Opus tier, a one-million-token context window available in beta. The update is notable not just for its technical enhancements but for its performance on several industry benchmarks, where Anthropic claims it now leads competitors, including OpenAI's GPT-5.2, in evaluations measuring performance on economically valuable professional tasks.

According to the company's release materials, Opus 4.6 surpasses GPT-5.2 by 144 Elo points on the GDPval-AA benchmark, which assesses knowledge work in fields like finance and law. The model also scored highest on the agentic coding test Terminal-Bench 2.0 and the information retrieval benchmark BrowseComp. Anthropic is keeping the API pricing for its top-tier model unchanged at $5 per million input tokens and $25 per million output tokens. New developer-focused features include 'adaptive thinking' to adjust cognitive effort based on task complexity, explicit 'effort controls' to manage the balance between performance and cost, and context compaction for longer-running agentic tasks. The model is available immediately through the claude.ai interface, the API, and major cloud service providers.

This release positions Anthropic to compete more directly in the enterprise market, where reliable performance on complex, long-duration tasks is critical. By emphasizing improved long-context retrieval and the ability to operate autonomously within large codebases and complex documents, the company is targeting professional workflows in software development, legal analysis, and finance. The introduction of features like agent teams and integrations with standard business software such as Excel and PowerPoint signals a strategic focus on moving beyond single-turn interactions toward building AI systems that function as persistent, capable collaborators within existing enterprise ecosystems.

By holding pricing steady while delivering benchmark-topping performance and a massive context window, Anthropic is aggressively challenging competitors on both capability and value, aiming to make its top-tier model the default choice for developers building complex, high-stakes enterprise agents.