Claude Sonnet 3.7

Claude 3.7 Sonnet is Anthropic’s latest Sonnet model, combining responsiveness and depth. In its standard mode, it answers quickly and cleanly—good for most prompts. But when you need to push on harder problems—math, coding, detailed analysis—it can shift into “extended thinking” mode, reflecting internally before producing its answer.

API users have control over how long it thinks (how many “thinking tokens” it uses), allowing trade-offs between speed and output quality. The model handles large contexts and delivers strong performance in end-to-end developer workflows, web development, instruction tasks, and scenarios that benefit from visible reasoning. In both modes, it supports function or tool calling and structured JSON outputs, making integration into agent pipelines, RAG systems, or engineering tasks smooth.

It’s available across Anthropic’s product tiers (Pro, Team, Enterprise) and on partner platforms such as Amazon Bedrock and Google Cloud’s Vertex AI.