Claude Sonnet 4 now supports 1M tokens of context (anthropic.com)

Article summary

Claude Sonnet 4 now supports up to 1 million tokens of context, a 5x increase that allows for processing entire codebases, synthesizing extensive document sets, and building agents that maintain coherence across hundreds of tool calls. This feature is available in public beta on the Claude Developer Platform and in Amazon Bedrock and Google Cloud's Vertex AI. The increased context support enables more comprehensive and data-intensive use cases, including large-scale code analysis and document synthesis. Pricing adjusts for prompts over 200K tokens to account for increased computational requirements.

Main themes

• Claude Sonnet 4 update
• Increased context support
• Code analysis and synthesis
• AI model limitations
• Pricing and cost savings
• Developer workflow and tools

What commenters say

• The increased context support in Claude Sonnet 4 may lead to decreased output quality due to the model getting 'distracted' by excessive context.
• Breaking down tasks into smaller, manageable chunks with fresh contexts can help mitigate the issue of decreased output quality.
• Some users find that Claude Code is more consistently productive, while others prefer Gemini Pro 2.5 for its ability to handle larger context windows.
• The choice of AI model depends on the specific use case, with some models exceling at coding and others at general queries or reasoning.
• Using multiple models in conjunction, such as Claude and Gemini Pro 2.5, can provide a more comprehensive workflow.
• The ability to switch between models and configure context settings is essential for effective use of AI-powered coding tools.
• The cost of using AI-powered coding tools can be significant, and users must weigh the benefits against the costs.
• The development of AI-powered coding tools is ongoing, and users can expect new features and improvements to be released regularly.