Anthropic released Claude Opus 4.6 on February 5, 2026, upgrading its flagship model with enhanced coding capabilities and a 1-million token context window in beta. The new model “plans more carefully, sustains agentic tasks for longer, can operate more reliably in larger codebases, and has better code review and debugging skills to catch its own mistakes,” according to Anthropic’s announcement.
The release marks the first time an Opus-class model has featured the expanded context window, allowing the model to process significantly more information in a single session. Opus 4.6 can apply these capabilities to “running financial analyses, doing research, and using and creating documents, spreadsheets, and presentations.”
Anthropic reports state-of-the-art performance across multiple benchmarks. The model achieved the highest score on Terminal-Bench 2.0, an agentic coding evaluation, and leads other frontier models on Humanity’s Last Exam, described as “a complex multidisciplinary reasoning test.” On GDPval-AA, which measures performance on economically valuable knowledge work tasks, Opus 4.6 outperformed OpenAI’s GPT-5.2 by approximately 144 Elo points and its predecessor Claude Opus 4.5 by 190 points.
The model introduces several new features for developers and enterprise users. Within Anthropic’s Cowork platform, Opus 4.6 can multitask autonomously. The API now supports “compaction” to summarize context for longer-running tasks and “adaptive thinking” where the model adjusts its reasoning depth based on contextual clues. Developers can control performance trade-offs through new effort controls using the /effort parameter.
Claude Opus 4.6 is available immediately on claude.ai, Anthropic’s API using the identifier “claude-opus-4-6,” and major cloud platforms. Pricing remains unchanged at $5/$25 per million tokens. Anthropic has also upgraded Claude in Excel and released Claude in PowerPoint as a research preview.
