claude-sonnet-4

completions

byanthropic

claude-sonnet-4

completions

Published by: anthropicProvider:

anthropic

Our high-performance model with exceptional reasoning and efficiency

Released

May 22, 2025

Knowledge

Mar 1, 2025

License

Proprietary

Context

200K

Input

$3 / 1M tokens

Output

$15 / 1M tokens

Cached

$0.3 / 1M tokens

Cached Write

$3.75 / 1M tokens

Capabilities: tools

Accepts: text, image

Returns: text

Released May 22, 2025Knowledge Cutoff: Mar 1, 2025License: Proprietary

Context: 200K Input: $3 / 1M tokensOutput: $15 / 1M tokensCached: $0.3 / 1M tokensCached Write: $3.75 / 1M tokensCapabilities: toolsAccepts: text, imageReturns: text

Access claude-sonnet-4 through LangDB AI Gateway

Recommended

Integrate with anthropic's claude-sonnet-4 and 250+ other models through a unified API. Monitor usage, control costs, and enhance security.

Unified API

Cost Optimization

Enterprise Security

Get Started Now

Free tier available • No credit card required

Instant Setup

99.9% Uptime

10,000+Monthly Requests

Benchmark Results for claude-sonnet-4

Overall Performance: 54.596250000000005 average score across all categories

Detailed Benchmark Scores:

Benchmark	Score	Percentile	Domain
AIME	40.70	Top 30%	Mathematics
AA Coding Index	41.10	Top 27%	Programming
AAII	44.40	Top 21%	General
AA Math Index	38.00	Top 46%	Mathematics
GPQA	75.40	Top 17%	STEM (Physics, Chemistry, Biology)
HLE	4.00	Top 77%	General Knowledge
LiveCodeBench	44.90	Top 30%	Programming
MATH-500	93.40	Top 24%	Mathematics
MMLU-Pro	83.70	Top 11%	General Knowledge
MMMU	74.40	Top 31%	General Knowledge
SciCode	37.30	Top 11%	Scientific

GPQA Score: 75.40 - Graduate-level reasoning benchmark

Model Comparison:

Provider: anthropic

Model Type: completions

Context Size: 200000 tokens

Comparing against 345 models in the database

Performance Analytics for claude-sonnet-4

Usage Statistics (Last 5 Days):

Total Requests: 31 API calls
Average TPS: 519.75 tokens per second
Average Response Time: 7982.60ms
Average Time to First Token: 1766.90ms
Total Cost: $0.55
Average Request Cost: $0.0177

Daily Performance Breakdown:

Date	Requests	TPS	Response Time	TTFT	Cost
9/14/2025	5	47.85	2591.70ms	1046.40ms	$0.01
9/15/2025	3	11.63	2407.60ms	2407.60ms	$0.00
9/16/2025	1	13.16	1368.00ms	1368.00ms	$0.00
9/17/2025	7	1688.47	3248.10ms	1352.50ms	$0.15
9/18/2025	15	440.53	13545.10ms	2098.90ms	$0.39

Performance Summary:

Model: claude-sonnet-4 by anthropic

Monitoring Period: 9/13/2025 to 9/20/2025

Average Daily Requests: 6

Peak Daily Requests: 15

Statistics

View Full Statistics

Request Volume

Daily API requests

Performance (TPS)

Tokens per second

519.75 tokens/s

Category Scores

vision#9

Benchmark Tests

View Other Benchmarks

AIME

40.7

Mathematics

AA Coding Index

41.1

Programming

AAII

44.4

General

AA Math Index

38.0

Mathematics

GPQA

75.4

STEM (Physics, Chemistry, Biology)

HLE

4.0

General Knowledge

LiveCodeBench

44.9

Programming

MATH-500

93.4

Mathematics

MMLU-Pro

83.7

General Knowledge

MMMU

74.4

General Knowledge

SciCode

37.3

Scientific

Metric	AIME	AA Coding Index	AAII	AA Math Index	GPQA	HLE	LiveCodeBench	MATH-500	MMLU-Pro	MMMU	SciCode
Score	40.7	41.1	44.4	38.0	75.4	4.0	44.9	93.4	83.7	74.4	37.3

Compare with Similar Models

gemini-2.5-pro

gpt-4.1

gemini-2.5-pro-preview

grok-4

gemini-2.5-flash

gemini-2.5-flash-preview

Code Examples

Integration samples and API usage

Code Samples for claude-sonnet-4

Python SDK Example:

from openai import OpenAI

client = OpenAI(
    base_url="https://api.us-east-1.langdb.ai/projects/<your_project_id>",
    api_key="<your_api_key>"
)

response = client.chat.completions.create(
    model="claude-sonnet-4",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ]
)

print(response.choices[0].message.content)

TypeScript SDK Example:

import OpenAI from 'openai';

const client = new OpenAI({
    baseURL: "https://api.us-east-1.langdb.ai/projects/<your_project_id>",
    apiKey: "<your_api_key>"
});

const response = await client.chat.completions.create({
    model: "claude-sonnet-4",
    messages: [
        { role: "user", content: "Hello, how are you?" }
    ]
});

console.log(response.choices[0].message.content);

cURL Example:

curl -X POST "https://api.us-east-1.langdb.ai/projects/<your_project_id>/v1/chat/completions" \
  -H "Authorization: Bearer <your_api_key>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4",
    "messages": [
        {"role": "user", "content": "Hello, how are you?"}
    ]
  }'

Model: claude-sonnet-4

Provider: anthropic

API Endpoint: $https://api.us-east-1.langdb.ai

Create API Key

Related Models

Similar models from anthropic

claude-opus-4

claude-3.7-sonnet

claude-3-5-sonnet-20240620

claude-3.5-haiku

claude-3-haiku-20240307

claude-3-opus-20240229

Conversations

Explore real conversations using this model

claude-sonnet-4 by anthropic - AI Model Details, Pricing, and Performance Metrics

claude-sonnet-4

claude-sonnet-4

Access claude-sonnet-4 through LangDB AI Gateway

Statistics

Category Scores

Benchmark Tests

Compare with Similar Models

Code Examples

Related Models

claude-opus-4

claude-3.7-sonnet

claude-3-5-sonnet-20240620

claude-3.5-haiku

claude-3-haiku-20240307

claude-3-opus-20240229

Conversations

Travel Planning Agent built with OpenAI Agents SDK

Web Search Agent built wtih Google ADK

HTML5 Canvas Freehand Drawing with Clear Button

Responsive 2-Column Photo Gallery with Hover Effect

Real-Time Username & Email Validation Form

Responsive Navbar with Hamburger Menu <600px

Pure HTML/CSS Image Slider with Fade & Arrows

pytest unit tests for is_palindrome with parametrize

A* Pathfinding on Grid (Python)

Two Sum: Python Efficient Solution

Flask To-Do List API with CRUD endpoints

Python Pygame Tetris Game with Classic Features

Closed-form for aₙ and exact value of a₅

Cylinder Surface Area Calculation

Probability both cards are face cards

Integer Solutions to 3x + 5y = 100

Bob wins when n ≡ 1, 3, or 4 (mod 5); count of such n ≤ 2024 is **1215**.

Origin & Meaning of "Ride Shotgun"

Synonyms and Nuances of "Gregarious"

Meaning and Usage of “Ubiquitous”

Counting “issi” occurrences and overlaps in sentence

Counting vowels in "encyclopedia"

Bridge Crossing Puzzle: Minimum Time Plan

3-Weighing Strategy to Find 1 Lighter Fake Coin

Meeting time & place of two trains on same track

Emily and Derek's Ages Algebra Problem

Andrew's Number of Sisters

Bob wins when n ≡ 1, 3, or 4 (mod 5); count of such n ≤ 2024 is 1215.