Anthropic Claude Models on LangMart

Overview

This document provides comprehensive details about all Anthropic Claude models available through LangMart. The models span multiple generations and are optimized for different use cases, from high-speed inference to deep reasoning tasks.

Note: The specific model anthropic/claude-web-research referenced in the request does not exist on LangMart. Below is the complete catalog of available Claude variants.

Model Families

Claude models are organized into several families:

Opus: Frontier reasoning models for complex tasks
Sonnet: Balanced models for production workloads
Haiku: Lightweight, fast models for real-time applications
Legacy Models: Older versions maintained for compatibility

Latest Generation Models (Recommended)

1. Claude Opus 4.5

Inference Model ID: anthropic/claude-opus-4.5

Description: Claude Opus 4.5 is Anthropic's frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use. It offers strong multimodal capabilities, competitive performance across real-world coding and reasoning benchmarks, and improved robustness to prompt injection. The model is designed to operate efficiently across varied effort levels, enabling developers to trade off speed, depth, and token usage depending on task requirements.

Key Features:

Frontier-level reasoning and problem-solving
Advanced tool use and multi-agent coordination
Strong performance on complex coding tasks
Improved robustness to prompt injection attacks
Extended context management
Controllable token efficiency via verbosity parameter

Pricing (per token):

Input: $0.000005 (5 µ per token)
Output: $0.000025 (25 µ per token)
Web Search: $0.01 per request
Input Cache Read: $0.0000005 (0.5 µ per token)
Input Cache Write: $0.00000625 (6.25 µ per token)

Specifications:

Context Window: 200,000 tokens
Input Modalities: Text, Image, Audio, Video
Output Modalities: Text
Maximum Completion Tokens: Varies

Supported Parameters:

include_reasoning, max_tokens, reasoning, response_format, stop,
structured_outputs, temperature, tool_choice, tools, top_k, verbosity

Use Cases:

Complex software engineering tasks
Agentic workflows requiring sustained reasoning
Long-horizon computer use automation
Multi-step planning and debugging
Spreadsheet and browser manipulation
Autonomous research tasks

Default Parameters:

Temperature: Default (1)
Top P: Default (0.9)

2. Claude Haiku 4.5

Inference Model ID: anthropic/claude-haiku-4.5

Description: Claude Haiku 4.5 is Anthropic's fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4's performance across reasoning, coding, and computer-use tasks, Haiku 4.5 brings frontier-level capability to real-time and high-volume applications.

It introduces extended thinking to the Haiku line, enabling controllable reasoning depth, summarized or interleaved thought output, and tool-assisted workflows with full support for coding, bash, web search, and computer-use tools. Scoring >73% on SWE-bench Verified, Haiku 4.5 ranks among the world's best coding models.

Key Features:

Fastest model in Claude lineup
Extended thinking support with controllable depth
Near-frontier reasoning capability
Exceptional coding performance (>73% on SWE-bench Verified)
Support for all tool types: coding, bash, web search, computer use
Highly cost-effective

Pricing (per token):

Input: $0.000001 (1 µ per token)
Output: $0.000005 (5 µ per token)
Input Cache Read: $0.0000001 (0.1 µ per token)
Input Cache Write: $0.00000125 (1.25 µ per token)

Specifications:

Context Window: 200,000 tokens
Input Modalities: Text, Image, Audio, Video
Output Modalities: Text
Maximum Completion Tokens: Varies

Supported Parameters:

include_reasoning, max_tokens, reasoning, stop, temperature,
tool_choice, tools, top_k, top_p

Use Cases:

High-volume inference workloads
Real-time chatbots and assistants
Sub-agent workflows in agentic systems
Parallelized task execution
Code completion and suggestions
Streaming applications requiring low latency
Cost-sensitive production deployments

Default Parameters:

Temperature: 1
Top P: 0.9

3. Claude Sonnet 4.5

Inference Model ID: anthropic/claude-sonnet-4.5

Description: Claude Sonnet 4.5 is Anthropic's most advanced Sonnet model to date, optimized for real-world agents and coding workflows. It delivers state-of-the-art performance on coding benchmarks such as SWE-bench Verified, with improvements across system design, code security, and specification adherence. The model is designed for extended autonomous operation, maintaining task continuity across sessions and providing fact-based progress tracking.

Sonnet 4.5 also introduces stronger agentic capabilities, including improved tool orchestration, speculative parallel execution, and more efficient context and memory management.

Key Features:

State-of-the-art coding performance
Extended autonomous operation
Improved tool orchestration
Speculative parallel execution
Efficient context/memory management
Fact-based progress tracking
Enhanced multi-context awareness

Pricing (per token):

Input: $0.000003 (3 µ per token)
Output: $0.000015 (15 µ per token)
Input Cache Read: $0.0000003 (0.3 µ per token)
Input Cache Write: $0.00000375 (3.75 µ per token)

Specifications:

Context Window: 1,000,000 tokens (1M)
Input Modalities: Text, Image, Audio, Video
Output Modalities: Text
Maximum Completion Tokens: Varies

Supported Parameters:

include_reasoning, max_tokens, reasoning, response_format, stop,
structured_outputs, temperature, tool_choice, tools, top_k, top_p

Use Cases:

Long-running agent workflows
Complex multi-step coding tasks
System design and architecture
Code security analysis
Multi-context document processing
Research automation
Financial analysis agents
Software engineering teams

Default Parameters:

Temperature: 1
Top P: 0.9

Production Models (Previous Generation)

4. Claude Opus 4.1

Inference Model ID: anthropic/claude-opus-4.1

Description: Claude Opus 4.1 is an updated version of Anthropic's flagship model, offering improved performance in coding, reasoning, and agentic tasks. It achieves 74.5% on SWE-bench Verified and shows notable gains in multi-file code refactoring, debugging precision, and detail-oriented reasoning. The model supports extended thinking up to 64K tokens.

Key Features:

Improved coding and reasoning
Extended thinking (up to 64K tokens)
Multi-file code refactoring
Precise debugging capabilities
Stable performance on complex tasks

Pricing (per token):

Input: $0.000015 (15 µ per token)
Output: $0.000075 (75 µ per token)
Image Input: $0.024 per image
Input Cache Read: $0.0000015 (1.5 µ per token)
Input Cache Write: $0.00001875 (18.75 µ per token)

Specifications:

Context Window: 200,000 tokens
Input Modalities: Text, Image
Output Modalities: Text
Extended Thinking: Up to 64K tokens

Supported Parameters:

include_reasoning, max_tokens, reasoning, response_format, stop,
structured_outputs, temperature, tool_choice, tools, top_k, top_p

5. Claude Opus 4 (Legacy)

Inference Model ID: anthropic/claude-opus-4

Description: Claude Opus 4 is benchmarked as the world's best coding model at its time of release. It brings sustained performance on complex, long-running tasks and agent workflows. It sets benchmarks in software engineering, achieving leading results on SWE-bench (72.5%) and Terminal-bench (43.2%).

Specifications:

Context Window: 200,000 tokens
Pricing: Input $0.000015, Output $0.000075 per token

6. Claude Sonnet 4

Inference Model ID: anthropic/claude-sonnet-4

Description: 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%), Sonnet 4 balances capability and computational efficiency.

Specifications:

Context Window: 1,000,000 tokens (1M)
Pricing: Input $0.000003, Output $0.000015 per token

Thinking / Extended Reasoning Models

7. Claude 3.7 Sonnet (thinking)

Inference Model ID: anthropic/claude-3.7-sonnet:thinking

Description: Claude 3.7 Sonnet with extended thinking mode enabled. This variant supports the reasoning parameter for step-by-step processing on complex tasks.

Key Features:

Extended thinking for complex reasoning
Hybrid reasoning approach
Improved coding performance
Better instruction following

Pricing (per token):

Input: $0.000003 (3 µ per token)
Output: $0.000015 (15 µ per token)
Image Input: $0.0048 per image

Specifications:

Context Window: 200,000 tokens

Supported Parameters:

include_reasoning, max_tokens, reasoning, stop, temperature,
tool_choice, tools, top_p

8. Claude 3.7 Sonnet (standard)

Inference Model ID: anthropic/claude-3.7-sonnet

Description: Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities. It demonstrates notable improvements in coding, particularly in front-end development and full-stack updates, and excels in agentic workflows.

Specifications:

Context Window: 200,000 tokens
Pricing: Input $0.000003, Output $0.000015 per token

Fast/Efficient Models

9. Claude 3.5 Sonnet

Inference Model ID: anthropic/claude-3.5-sonnet

Description: New Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices. Sonnet is particularly good at:

Coding: Scores ~49% on SWE-Bench Verified (higher than previous best)
Data Science: Augments human expertise; navigates unstructured data
Visual Processing: Excels at interpreting charts, graphs, and images
Agentic Tasks: Exceptional tool use for complex, multi-step problem solving

Specifications:

Context Window: 200,000 tokens
Pricing: Input $0.000006, Output $0.00003 per token

Supported Parameters:

max_tokens, stop, temperature, tool_choice, tools, top_k, top_p

10. Claude 3.5 Haiku

Inference Model ID: anthropic/claude-3.5-haiku or anthropic/claude-3.5-haiku-20241022

Description: Claude 3.5 Haiku features enhancements across all skill sets including coding, tool use, and reasoning. As the fastest model in the Anthropic lineup, it offers rapid response times suitable for high-interactivity applications.

Excels in:

Real-time chatbots
On-the-fly code completions
Data extraction
Real-time content moderation

Note: The specific version 20241022 does not support image inputs.

Specifications:

Context Window: 200,000 tokens
Pricing: Input $0.0000008, Output $0.000004 per token
Image Input: Not supported (20241022 variant)

Supported Parameters:

max_tokens, stop, temperature, tool_choice, tools, top_k, top_p

11. Claude 3 Haiku

Inference Model ID: anthropic/claude-3-haiku

Description: Claude 3 Haiku is Anthropic's fastest and most compact model for near-instant responsiveness. Quick and accurate targeted performance with multimodal support.

Specifications:

Context Window: 200,000 tokens
Pricing: Input $0.00000025, Output $0.00000125 per token
Image Input: $0.0004 per image

Legacy/Older Models

12. Claude 3 Opus

Inference Model ID: anthropic/claude-3-opus

Description: Claude 3 Opus is Anthropic's most powerful model for highly complex tasks. It boasts top-level performance, intelligence, fluency, and understanding.

Specifications:

Context Window: 200,000 tokens
Pricing: Input $0.000015, Output $0.000075 per token
Image Input: $0.024 per image

Pricing Summary Table

Model	Input (per token)	Output (per token)	Context	Best For
Opus 4.5	$0.000005	$0.000025	200K	Frontier reasoning, complex tasks
Sonnet 4.5	$0.000003	$0.000015	1M	Production agents, coding workflows
Haiku 4.5	$0.000001	$0.000005	200K	Real-time, high-volume, cost-sensitive
Opus 4.1	$0.000015	$0.000075	200K	Complex reasoning tasks
Sonnet 4	$0.000003	$0.000015	1M	Production workloads
3.5 Sonnet	$0.000006	$0.00003	200K	Balanced performance
3.5 Haiku	$0.0000008	$0.000004	200K	Speed-optimized inference
3 Haiku	$0.00000025	$0.00000125	200K	Legacy, ultra-fast
3 Opus	$0.000015	$0.000075	200K	Legacy, maximum capability

Feature Comparison

Extended Thinking / Reasoning Support

Models with native extended thinking support:

Claude Opus 4.5 (via reasoning parameter)
Claude Haiku 4.5 (via reasoning parameter)
Claude Sonnet 4.5 (via reasoning parameter)
Claude Opus 4.1 (via reasoning parameter)
Claude Opus 4 (via reasoning parameter)
Claude Sonnet 4 (via reasoning parameter)
Claude 3.7 Sonnet (via reasoning parameter)
Claude 3.7 Sonnet (thinking) (always enabled)

Vision / Multimodal Support

Models with image input support:

Claude Opus 4.5
Claude Haiku 4.5
Claude Sonnet 4.5
Claude Opus 4.1
Claude Sonnet 4
Claude 3.7 Sonnet
Claude 3.5 Sonnet (text only)
Claude 3 Haiku
Claude 3 Opus

Note: Claude 3.5 Haiku (20241022) does not support images.

Tool Use & Function Calling

All Claude models support tool use with:

tool_choice parameter (force tool use)
tools parameter (pass tool definitions)
Streaming tool calls
Parallel tool execution (Sonnet 4.5+)

API Usage Examples

Basic Chat Completion

curl -X POST https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer $LANGMART_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-opus-4.5",
    "messages": [
      {"role": "user", "content": "Explain quantum computing"}
    ],
    "temperature": 0.7,
    "max_tokens": 1024
  }'

Extended Thinking (Reasoning)

curl -X POST https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer $LANGMART_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-opus-4.5",
    "messages": [
      {"role": "user", "content": "Solve this math problem: 17 × 23"}
    ],
    "reasoning": "enabled",
    "include_reasoning": true,
    "max_tokens": 8000
  }'

With Tools

curl -X POST https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer $LANGMART_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-4.5",
    "messages": [
      {"role": "user", "content": "What is the weather in New York?"}
    ],
    "tools": [
      {
        "type": "function",
        "function": {
          "name": "get_weather",
          "description": "Get weather for a location",
          "parameters": {
            "type": "object",
            "properties": {
              "location": {"type": "string"}
            }
          }
        }
      }
    ],
    "tool_choice": "auto"
  }'

Vision/Image Input

curl -X POST https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer $LANGMART_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-opus-4.5",
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "type": "image_url",
            "image_url": {
              "url": "https://example.com/image.jpg"
            }
          },
          {
            "type": "text",
            "text": "Describe this image"
          }
        ]
      }
    ]
  }'

Context Caching (Sonnet 4.5 with 1M context)

curl -X POST https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer $LANGMART_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-4.5",
    "messages": [
      {
        "role": "user",
        "content": "Analyze this large document..."
      }
    ],
    "max_tokens": 1024
  }'

Model Selection Guide

Choose Opus 4.5 if you need:

Maximum reasoning capability
Complex software engineering tasks
Long-horizon automation
Frontier-level intelligence
Best-in-class performance regardless of cost

Choose Sonnet 4.5 if you need:

Production-grade performance
Extended context (1M tokens)
Balanced cost and capability
Long-running agent workflows
State-of-the-art coding performance

Choose Haiku 4.5 if you need:

Lowest latency
Cost-sensitive applications
High-volume inference
Real-time interactions
Fast sub-agents in larger systems

Choose Legacy Models if:

You have existing integrations
You need specific benchmark results
You're testing backward compatibility
You prefer older, stable APIs

Limitations & Considerations

Claude-Web-Research: This specific model variant does not exist on LangMart. Use Claude Opus 4.5 or Sonnet 4.5 with web search capability if needed.
Rate Limiting: LangMart may apply rate limits based on your plan.
Extended Thinking Costs: Using reasoning: "enabled" incurs additional costs and increases latency.
Context Window: Different models have different context windows (200K or 1M). Larger contexts cost more.
Image Input Costs: Image inputs are charged at $0.0048 (Sonnet 4) to $0.024 (Opus 4) per image.
Deprecated Models: Older Claude 3 models should not be used for new projects.

Web Search Capability

Several Claude models support web search (billed at $0.01 per request):

Claude Opus 4.5
Claude Opus 4.1
Claude Opus 4
Claude 3.7 Sonnet
Claude 3.7 Sonnet (thinking)

Use by adding web_search parameter in your requests.

References

LangMart API Documentation: https://langmart.ai/docs
Anthropic Official Site: https://www.anthropic.com
LangMart Models: https://langmart.ai/model-docs

Last Updated: December 23, 2025 Data Source: LangMart API v1/models endpoint Status: All information current and verified