Claude 4 sonnet
Anthropic's hybrid reasoning model for
high-end coding and reasoning use cases
CompareTry with Bind AI
Reasoning
✩✩✩✩✩
Excellent
Pricing
$3/$15
Input/Output per million tokens
Input
Text, Image
Output
Text, Code Execution
Speed
Medium

Claude 4 Sonnet Overview

Claude 4 Sonnet is Anthropic's practical workhorse—offering upgraded capabilities at the same price as its predecessor, Claude 3.7 Sonnet (try it here). Optimized for both everyday development tasks, such as code reviews and bug fixes, as well as advanced code-generation tasks, Claude 4 Sonnet excels and impresses as a coding-first model. It supports text, images, and files as input and features a 200K token context window.

‍

❝Claude 4 Sonnet offers affordable pricing with free access (limited) and via paid plans: Pro at $17 per month and Max at $100 per month. Features like batch processing and prompt caching can reduce costs by up to 90%, optimizing high-volume workflows.❞

‍

Learn more about Claude 4 Sonnet and how it compares to Claude 3.7 Sonnet and Gemini 2.5 Pro here.

‍

Claude 4 Sonnet Pricing

Claude 4 Sonnet offers competitive pricing that is comparable to that of its predecessor.

‍

Input

$3 / MTok

‍

Prompt caching write

$3.75 / MTok

‍

Prompt caching read

$0.30 / MTok

‍

Output

$15 / MTok

‍

Claude 4 Sonnet Context Window

Claude 4 Sonnet's context window remains unchanged from its predecessor, at 200,000. It's on par with OpenAI's flagship reasoning model, OpenAI o3, but falls short of Gemini 2.5 Pro's exceptional 1,000,000 context window.

‍

Context Window Comparison
200K
Claude Sonnet 4
200K
Claude 3.7 Sonnet
200K
OpenAI o3
1M
Gemini 2.5 Pro
Claude Sonnet 4
Claude 3.7 Sonnet
OpenAI o3
Gemini 2.5 Pro

Claude 4 Sonnet Benchmarks

‍

Claude 4 Sonnet sets an impressive standard for AI-powered coding and reasoning. The table below summarizes its standout features and technical capabilities.

‍

Benchmark Score Notes
MMLU 88.0% Outstanding performance across 57 academic subjects.
HumanEval (Coding) 73% Solid code generation and reasoning under constraints.
GSM8K 95% High accuracy in multi-step math problems.
HellaSwag 95.6% Strong commonsense reasoning.
ARC (Challenge) 71% Improved performance over earlier versions.
TriviaQA 93% Excellent factual recall and comprehension.
Winogrande 90.2% Consistent logical disambiguation and pronoun resolution.
BIG-Bench Hard 65% Good performance on difficult reasoning tasks.

Try Claude 4 Sonnet with these prompts

Create a modern microblogging platform with real-time feed, tweets, likes, and retweets functionality.

Try This Prompt

Airbnb Clone

Build a property rental marketplace with listings, search filters, booking system, and user reviews.

Try This Prompt

Spotify Clone

Develop a music streaming app with playlists, a responsive player, and search for tracks and artists.

Try This Prompt

Instagram Clone

Create a photo-sharing social network with grid layout, likes, comments, and stories.

Try This Prompt
One Platform, Every Model

Access all top AI models with Bind AI

One subscription unlocks Claude 4, OpenAI o4-mini, Gemini 2.5 Pro, DeepSeek R1, and more powerful models

Claude 4
OpenAI o4-mini
Gemini 2.5 Pro
DeepSeek R1
+ More
✓
GitHub Integration
✓
Custom GPTs
✓
Premium Features
Start creating with Bind AI →