Sponsored by Byond Boundrys Consulting - Empowering Ideas, Delivering Results
Anthropic Advanced

Claude Sonnet 5

Claude Sonnet 5 is Anthropic's active 1M-context agentic model with up to 128K output, January 2026 knowledge cutoff and $2/M input pricing.

General Purpose Language ModelTextImage Paid
In plain English

What is this model and why does it matter?

Claude Sonnet 5 is Anthropic's active 1M-context agentic Sonnet model for coding, tool use and professional knowledge work.

Coding agentsProfessional knowledge workLong-horizon tasksComputer useTool useLarge-context analysis
Model overview

Claude Sonnet 5: features, use cases and important details

Claude Sonnet 5 is Anthropic’s current Sonnet model, released June 30, 2026.

Verified model facts

Anthropic documents a 1M-token context window, up to 128K output, January 2026 knowledge cutoff, text and image input, adaptive reasoning, tools and advanced agentic coding capabilities.

Current status

It is active on the Claude API and multiple cloud platforms, priced at $2/M input and $10/M output.

Best fit

It is best for coding agents, professional knowledge work, long-horizon tasks and very large-context document workflows.

Limitations

It remains a proprietary model, static knowledge should be augmented with tools when freshness matters, and large agent loops can generate substantial token usage.

Get started

How to use this model

  1. Create an Anthropic API key.
  2. Call the Messages API with model claude-sonnet-5.
  3. Use adaptive thinking and effort controls for harder tasks.
  4. Add tools/browser/computer-use capabilities where appropriate.
  5. Monitor the new tokenizer's token counts and 1M-context cost.
Copy and try

Example prompts

  • Plan and complete this multi-file software migration using the available tools.
  • Analyze these long documents and produce an actionable decision memo.
  • Use browser and terminal tools to complete this operational task.
Capabilities

What it can do

  • 1M context
  • 128K max output
  • Adaptive reasoning
  • Coding
  • Tool use
  • Computer use
  • Vision
  • Agentic workflows
Best for

Practical use cases

  • Software engineering
  • Agents
  • Research
  • Knowledge work
  • Long documents
Pricing

What does it cost?

$2.00 per 1M input tokens and $10.00 per 1M output tokens on the Claude API; Anthropic made this pricing permanent in August 2026.

Input$2.00 / 1M tokens
Output$10.00 / 1M tokens
Simple summaryCurrent Claude API pricing is $2/M input and $10/M output, with prompt caching and batch discounts available.

What stands out

  • 1M context
  • Strong agentic performance
  • Lower Sonnet pricing
  • Large 128K output limit
  • Broad platform availability

Things to consider

  • Proprietary
  • New tokenizer can increase token counts vs prior Sonnet
  • High-effort agent tasks can still be expensive
Limitations

Important restrictions and trade-offs

  • January 2026 static knowledge cutoff without tools
  • Can hallucinate or make agentic mistakes
  • Some tool/platform features differ by hosting provider
SimplifyAITools verdict

Our editorial take

A leading current general-purpose Claude model for coding and agents, offering unusually large context/output limits at Sonnet-class pricing.

References

Primary sources

  1. Open source 1 ↗
  2. Open source 2 ↗
  3. Open source 3 ↗
Keep exploring

Related AI models