Claude Sonnet 4 can now take up to 1 million context windows. This means users and developers could write prompts that are longer than books. Anthropic, the company behind Claude, has integrated this new update to attract more developer-centric businesses to their AI model.
This new token context window for Anthropic’s Claude is five times larger compared to its previous model and double that of ChatGPT-5’s window. This now lets you write 75,000 lines of code or 750,000 words, in a single prompt.
Anthropic has raised one of the largest enterprise businesses among AI model developers, mostly by selling their AI coding platform Claude to Windsurf, Microsoft’s GitHub Copilot and Anysphere’s Cursor. And for most developers, it seems to be their go-to AI model, but the new ChatGPT-5 may threaten them with its great subscription model and coding performance
The expanded context window for Claude allows developers to tackle more complex and data-heavy tasks. Anthropic said that this includes large-scale code analysis, where developers can load entire codebases and have Claude understand the project’s architecture, identify dependencies, and suggest improvements. This also enables document synthesis, allowing the processing of vast sets of documents, such as legal contracts or research papers, to analyse relationships between them while keeping a full understanding of the context.
Anthropic also claims that developers can build more sophisticated context-aware agents that can maintain a complete understanding of multi-step workflows, API documentation, and interaction histories, even across hundreds of tool calls.
On the contrary, OpenAI generates most of its revenue through the sale of its subscription to their AI model, ChatGPT, and Anthropic’s business is oriented towards selling AI models to business enterprises through an application platform interface.
Last week, Anthropic unveiled an updated version of its largest AI model, Claude Opus 4.1, which pushed the company’s AI coding capabilities a bit further.
Story continues below this ad
In broad terms, AI models that have more context perform better on all tasks, but particularly on software engineering challenges.
Additionally, Brad Abrams, the product lead for Claude, in an interview with TechCrunch, said that Claude’s broad context window improves its performance on lengthy agentic coding assignments, where the AI model works on an issue on its own for minutes or hours. Claude can recall every step it has taken in long-horizon activities when it has a broad context window.
Large context windows have been overdone by some businesses, who assert that their AI models are capable of processing enormous amounts of cues. For Gemini 2.5 Pro, Google provides a context window of two million tokens, while Meta provides a context window of ten million tokens for Llama 4 Scout.
To account for the extra computing power required for large requests, Anthropic will increase the cost for Claude Sonnet 4 API users when prompts exceed 200,000 tokens. The input token price will double from $3 to $6 per million tokens, and the output token price will increase from $15 to $22.50 per million tokens.
© IE Online Media Services Pvt Ltd