Claude Sonnet 4
by Anthropic
Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved precision and controllability. Ac
Models
Claude Sonnet 4
Added 10 July 2026
Overview
Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved precision and controllability. Achieving state-of-the-art performance on SWE-bench (72.7%),...
Best for
Best for
Production coding agents and long-context document work that need vision
Use cases
- repo agents
- code review
- document Q&A
- multimodal support bots
How to use / API access
Call via OpenRouter with model id `anthropic/claude-sonnet-4`.
- openrouter ·
anthropic/claude-sonnet-4Docs
OpenRouter id: anthropic/claude-sonnet-4
Benchmarks
- Intelligence Index (artificial-analysis) : 28.9 index
- Coding Index (artificial-analysis) : 37.6 index
- Agentic Index (artificial-analysis) : 16.6 index
- Design Arena 3d (other) : 1206 elo · #45
- Design Arena codecategories (other) : 1178 elo · #60
- Design Arena dataviz (other) : 1187 elo · #57
- Design Arena gamedev (other) : 1197 elo · #50
Enterprise DNA angle
Enterprise DNA runs Sonnet-class models inside Omni with secrets, jobs, and eval harnesses so coding agents ship without babysitting providers.
Notes
Claude Sonnet 4 is listed in the EDNA Models directory from the OpenRouter catalogue.
Pros
- Strong coding and agent loops with controllable outputs
- 1M context for large repos and long documents
- Vision + text in one model for multimodal workflows
Cons
- Output tokens cost more than open-weight alternatives
- Closed weights — no self-host option
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.