Claude Sonnet 4

by Anthropic

Claude Sonnet 4 is a significant upgrade to Claude Sonnet 3.7, reaching a state-of-the-art 72.7% on SWE-bench with stronger coding and reasoning. It balances capability with efficiency and responds more precisely to instructions for greater control over implementation. Sonnet 4 supports extended thinking with integrated tool use and parallel tool execution, enabling capable agentic workflows. It pairs vision with text analysis and shows substantial gains in multi-file development, with fewer navigation errors across a codebase. With a 200K token context window and support for tools, vision, and PDF processing, Sonnet 4 produces clean code with precise, surgical edits. It suits everyday applications that want near-frontier performance at practical efficiency.

Key info

Input
Output
Features
Context window
1M
Max output
64K

Benchmarks

Independent benchmark scores — composite indices for reasoning, coding, and math, plus individual eval scores where available.

Global rank#40 of 596 LLMs
TierStrong
Output speed0 tok/s
First token0.00s
Intelligence Index48.4
Coding Index63.0
Reasoning & knowledge
GPQA Diamond
88%
Humanity's Last Exam
34%
Long-context reasoning
74%
Coding
SciCode
47%
Agentic & tool use
Terminal-Bench Hard
53%
τ²-Bench Telecom
76%
Math & instruction following
IFBench
57%

Available routes

No routes currently available — Claude Sonnet 4 isn't routed through the Opper gateway right now. It may return.

Contact us about this model →

Available models from Anthropic

Start building with 700+ models

One API key. Every major provider. Up and running in minutes.

Get startedView Documentation
Claude Sonnet 4 by Anthropic — not currently on Opper | Opper AI