Anthropic has launched Claude Opus 4.6, its most advanced artificial intelligence model. The company says the model is better at coding, can handle longer and more complex tasks, and produces more professional work outputs.
The model release follows earlier updates last year, including Claude Opus 4.5, Sonnet 4.5, and Haiku 4.5.
What Is Anthropic?
Anthropic is a US-based artificial intelligence company founded in 2021 by former OpenAI researchers and executives. It is best known for its Claude family of AI models, which compete with models from OpenAI, Google, and others.
Anthropic positions Claude as an AI system focused on safety, reliability, and real-world work use.
Anthropic's Claude Model Family
Anthropic releases Claude models in three main sizes:
Opus - the largest and most powerful model
Sonnet - a mid-sized model for balanced performance
Haiku - a smaller, faster model for lightweight tasks
Claude Opus 4.6 is the most capable model in this lineup.
What's New In Claude Opus 4.6?
According to Anthropic, Claude Opus 4.6 builds on the strengths of earlier versions with clear improvements:
- Better planning and reasoning.
- Stronger coding, code review, and debugging skills
- More reliable performance in large codebases.
- Improved ability to work on long tasks without losing context.
For the first time in the Opus series, the model also supports a 1 million token context window in beta, allowing it to process extremely large documents or long conversations.
Features Of Anthropic Claude Opus 4.6
- Reviewing and debugging complex code.
- Working across large software projects.
- Pulling relevant information from long documents.
- Conducting research and financial analysis.
Claude Opus 4.6 currently leads the Finance Agent benchmark, which measures how well AI models perform core financial analyst tasks.
Better At Long Documents
One common issue with AI models is performance drop during long conversations or large inputs, often called “context rot.” Anthropic says Opus 4.6 shows major improvements here.
The model can track and retrieve information across hundreds of thousands of tokens with less drift, making it more useful for legal, financial, research, and enterprise work.
Designed For Knowledge Workers, Developers
Claude Opus 4.6 is built for everyday professional use. It can:
- Run financial models
- Assist with research
- Create and edit documents, spreadsheets, and presentations
- Work autonomously on longer tasks using Anthropic's Cowork environment
Anthropic has also upgraded Claude's integration with tools like Excel and introduced a research preview of Claude in PowerPoint.
New Features For Developers
Anthropic has added several updates to support developers using the Claude API:
- Adaptive thinking, where the model decides when deeper reasoning is needed.
- Effort controls to balance speed, cost, and intelligence.
- Context compaction, summarises older content to allow longer tasks.
- Support for up to 128,000 output tokens.
Pricing remains unchanged at $5 per million input tokens and $25 per million output tokens, with higher pricing for very large prompts.
Where Claude Opus 4.6 Is Available
Claude Opus 4.6 is available starting today through:
- claude.ai (Anthropic's chatbot interface)
- The Claude API
- All major cloud platforms
Developers can access it using the model name claude-opus-4-6.
Anthropic says Claude Opus 4.6 maintains a strong safety profile. In internal testing, the model showed low rates of harmful or misaligned behavior, such as deception or misuse. The company has also added new safeguards, especially around cybersecurity tasks, and conducted its most extensive safety testing so far.
