Claude Sonnet

Overview

This is a model family overview. For version-specific details, see the individual model entries linked below.

Claude Sonnet is the middle tier in Anthropic’s lineup and still the most practical starting point for most teams. Anthropic’s current documentation continues to frame Sonnet 4.6 as the balanced daily-use Claude tier for coding, analysis, writing, and tool-enabled assistants where quality matters but Opus pricing is hard to justify at scale.

Current Latest

As of April 18, 2026, Claude Sonnet 4.6 remains the current public Sonnet reference on Anthropic’s model page. Opus 4.7 now sits above it as the premium top-end Claude, but Sonnet 4.6 is still the practical default-tier route for most Claude-based work. This repo keeps Claude Sonnet 4.5 only for pinned deployments and historical comparison.

Strengths

• Strong general-purpose reasoning at a moderate price point
• Excellent coding assistance for everyday development tasks
• Good multimodal understanding for images and documents
• Strong instruction following for structured outputs and tool use
• Broad availability across Anthropic API, Bedrock, Vertex AI, Microsoft Foundry, and Claude products

When to Choose Sonnet

Sonnet is the default choice for most workloads:

• Day-to-day coding assistance, code review, and generation
• Content writing, analysis, and summarization
• Standard agentic workflows where Opus-level depth is not required
• Production systems where cost matters but quality cannot be sacrificed
• Interactive applications where response time matters

Access

• Anthropic API as claude-sonnet-4-6
• AWS Bedrock
• Google Vertex AI
• Microsoft Foundry
• Claude consumer and workspace products, including Claude Code and broader Claude desktop or workspace flows where available