Claude Sonnet 4

Model Overview

Description

Claude Sonnet 4 represents a significant upgrade from its predecessor, Claude Sonnet 3.7, with particular strength in coding and reasoning tasks. The model achieves state-of-the-art performance on SWE-bench with a score of 72.7%, demonstrating exceptional capability in software engineering tasks.

Key improvements include:

• Enhanced autonomous codebase navigation
• Reduced error rates in agent-driven workflows
• Increased reliability following complex instructions
• Improved precision and controllability
• Better balance between capability and computational efficiency

The model targets practical everyday applications with advanced reasoning while maintaining efficiency, making it suitable for both development and production workloads.

Pricing

Standard Pricing (Google Vertex Provider)

Price per Token (Detailed)

Capabilities

Supported Parameters

Best Practices

1. For Coding Tasks: Leverage the model's SWE-bench performance for code generation, review, and debugging
2. For Agent Workflows: Take advantage of improved reliability in multi-step autonomous tasks
3. For Long Context: Utilize the 1M token context window for large document analysis
4. For Cost Optimization: Use caching for repeated context to reduce input costs
5. For Complex Reasoning: Enable reasoning mode for step-by-step problem solving

API Usage Example

LangMart Format

curl https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer $LANGMART_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-4",
    "messages": [
      {"role": "user", "content": "Hello, Claude!"}
    ]
  }'

LangMart Format

curl -X POST https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-4",
    "messages": [
      {"role": "user", "content": "Hello, Claude!"}
    ],
    "max_tokens": 4096
  }'

With Reasoning Mode

curl -X POST https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-4",
    "messages": [
      {"role": "user", "content": "Solve this step by step: ..."}
    ],
    "reasoning": true,
    "include_reasoning": true
  }'

With Tool Calling

curl -X POST https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-4",
    "messages": [
      {"role": "user", "content": "What is the weather in Tokyo?"}
    ],
    "tools": [
      {
        "type": "function",
        "function": {
          "name": "get_weather",
          "description": "Get current weather for a location",
          "parameters": {
            "type": "object",
            "properties": {
              "location": {"type": "string"}
            },
            "required": ["location"]
          }
        }
      }
    ]
  }'

Related Models

Claude 4 Family

Previous Generations

Providers

Available Providers

Provider Order (Default)

1. Anthropic
2. Amazon Bedrock
3. Google Vertex

Supported Modalities

Input Modalities

• Text
• Images
• Files

Output Modalities

• Text only

Performance Metrics

Benchmarks

Usage Statistics (December 2025)

The model shows strong adoption with daily requests ranging from 326,000 to over 1 million.

Source

• LangMart Model Documentation: https://langmart.ai/model-docs
• Anthropic Documentation: https://docs.anthropic.com/
• Last Updated: December 23, 2025