Claude Opus 4.6: 1M context, agent teams, and leadership on Terminal-Bench 2.0
In one sentence Anthropic releases Opus 4.6: first Opus with 1M-token context in beta, agent teams in Claude Code, leadership on Terminal-Bench 2.0 and Humanity's Last Exam. Pricing unchanged at $5/$25.
Opus is Anthropic's "premium" model — the one you pull out when a task is genuinely hard. Version 4.6 lands on February 5, 2026 and brings three big things.
First: context goes up to 1 million tokens in beta. It's the first time an Opus-class model hits 1M (Sonnet already had it). It means you can feed it entire document archives, large codebases, months of conversations, without trimming.
Second: "agent teams" in Claude Code. Instead of a single agent working sequentially, you can now have a team of agents splitting the work: one researches, one writes, one tests. Same paradigm as a human team, automated.
Third: for agentic coding (Terminal-Bench 2.0) and multidisciplinary reasoning (Humanity's Last Exam) it's the strongest model on the market at release. Pricing unchanged from Opus 4.5: $5 input, $25 output per million tokens. Available on Claude.ai, API, Bedrock, Vertex AI.
Companies
Anthropic
Tools
Claude Opus 4.6, Claude Code, Claude Cowork
Tags
Sources