Anthropic’s Claude Opus 4.6 Debuts 1M-Token Context

Anthropic has released Claude Opus 4.6, a major update to its flagship AI model that the company says is better at planning, sustains longer autonomous workflows, and outperforms rivals like OpenAI’s GPT-5.2 on key enterprise benchmarks.

The rollout lands at a volatile time for the AI and software sectors. It follows just three days after OpenAI launched its Codex desktop app for macOS, positioning the two companies in more direct competition around developer tooling. It also comes against the backdrop of a $285 billion sell-off in software and services stocks, which some investors are linking to concerns that Anthropic’s AI systems could disrupt established enterprise software businesses.

With Opus 4.6, Anthropic is extending its top-tier Opus models to a 1 million token context window for the first time. That expanded context is designed to let the AI process and reason over far larger volumes of information than previous versions could handle in a single session.

Alongside the core model upgrade, Anthropic is adding a new capability inside Claude Code called “agent teams,” released as a research preview. The feature allows multiple AI agents to work at the same time on different parts of a coding project while coordinating with each other autonomously.

According to an Anthropic spokesperson speaking to VentureBeat, the company is “focused on building the most capable, reliable, and safe AI systems.” They said Opus 4.6 is “even better at planning, helping solve the most complex coding tasks.”

Describing how agent teams are intended to be used, the spokesperson said the feature lets users divide work across several specialized agents for example, one handling the frontend, another responsible for the API, and a third focused on a migration task with each “owning its piece and coordinating directly with the others.”

The Claude Opus 4.6 launch underscores how quickly the competition around AI for coding and enterprise workflows is intensifying. OpenAI’s recent Codex desktop application directly targets the same developer productivity space where Anthropic’s Claude Code has been gaining attention. At the same time, market worries about AI-driven disruption are feeding into broader volatility in software valuations, with Anthropic’s tools cited by some investors as a factor in a massive sector-wide rout.

Anthropic, for its part, is positioning Opus 4.6 as a step forward in careful planning, reliability, and long-running autonomous work and is betting that features like 1M-token context windows and coordinated agent teams will appeal to enterprises looking to tackle complex, code-heavy projects with AI assistance.