Back to models
LegacyChat Models

Anthropic

Claude Sonnet 5

Best Overall

Anthropic’s active legacy Sonnet model for coding and general work, with adaptive thinking and a 1M-token context window.

BalancedGeneral purpose
Open AppBrowse All Models

At a glance

Know the model before you prompt.

Anthropic API specifications
Context window
1,000,000 tokens
Maximum output
128,000 tokens
Inputs → output
Text + Images → Text
Knowledge cutoff
January 2026

API model ID: claude-sonnet-5

Capabilities & boundaries

What it supports. Where the limits are.

Tool support requires the appropriate API integration; a supported tool is not automatically active in every chat.

Supported API features and tools

  • Text and image input; text output
  • Adaptive thinking with configurable effort
  • Messages API tool-use workflows with an appropriate integration
  • Prompt caching with 5-minute and 1-hour write options
  • Asynchronous Message Batches processing

Before you choose

  • No native audio, video, or image output; accepting images does not make this an image generator.
  • This is an active legacy model, not the latest Sonnet release.
  • Non-default temperature, top_p, and top_k values return an error.
  • Sonnet 5.5 changes thinking and tool-use behavior; test a migration rather than carrying every request parameter forward.

Adaptive thinking and effort

lowmediumhigh · defaultxhighmax

Adaptive thinking is on by default with high effort. All five effort levels are available, and thinking can be turned off with disabled. Manual enabled/budget_tokens and between_tools are not supported. Reserve enough max_tokens for thinking plus the answer and use your own evaluations to choose effort.

  • Anthropic lists this model as active legacy; released June 30, 2026. Retirement is not sooner than June 30, 2027. This is an availability commitment, not a scheduled retirement.
  • Normal Messages output is limited to 128,000 tokens. Up to 300,000 output tokens are available only through the Message Batches API beta with the output-300k-2026-03-24 header.
  • The displayed cutoff is Anthropic’s reliable knowledge cutoff, published to month precision. A cutoff does not replace checking current facts or supplied evidence.
  • When moving to Sonnet 5.5, replace thinking-disabled configurations with supported between_tools/adaptive behavior and review forced-tool restrictions.
  • The documented API ID is claude-sonnet-5. Keep release and evaluation dates with your results to distinguish model generations.

Put it to work

Start with a more useful prompt.

Original examples from EZ Ai Assist. Adapt these to your task and the features available in your workspace.

Workflow 01

Review a routine pull request

Use the diff and tests as the review boundary.

Review this pull request for the behavior described in its issue. Focus on correctness, error handling, and missing tests within the changed lines and directly affected callers. For each finding, cite a code location and a failing scenario. Separate optional cleanup from defects. Do not claim the test suite passed unless the supplied output demonstrates it.

Workflow 02

Create a grounded internal answer

Answer from the supplied knowledge base with visible gaps.

Answer this employee question using only the knowledge-base excerpts below. Cite the section behind each instruction and preserve any eligibility conditions or exceptions. If the excerpts conflict or do not answer part of the question, explain the gap and identify the team that should clarify it only if named in the source. Do not invent policy.

Workflow 03

Audit a requirements checklist

Expose omissions and contradictions before implementation.

Compare this requirements checklist with the supplied customer notes. Mark each requirement as supported, contradicted, or not evidenced, and quote the short passage that justifies the classification. Identify duplicated items and undefined acceptance criteria. Finish with a prioritized list of clarification questions without silently adding new scope.

Developer reference

Anthropic API pricing

These are Anthropic API reference prices, not EZ Ai Assist subscription prices.

View EZ Ai Assist plans
Standard processing · USD per 1,000,000 tokens
Token typePrice
Input$2.00
5-minute cache write$2.50
1-hour cache write$4.00
Cache read$0.20
Output$10.00
  • Standard rates apply across the full 1M-token context window; no separate long-context surcharge is listed for this model.
  • Batch processing discounts input and output by 50%. Cache writes and reads have distinct rates; calculate from actual usage rather than applying a cache-read discount to the entire request.
  • Thinking tokens are billed as output even when their text is hidden. Tool charges and optional US-only inference (1.1× token pricing) can also affect the total.
  • The previously scheduled increase to $3/$15 did not occur. Sonnet 5’s $2 input / $10 output rates are now Standard, not an expired introductory offer.
  • Verify current provider pricing before estimating spend. These rates do not describe an EZ Ai Assist subscription.

Common questions

A few things worth knowing.

Is Sonnet 5 still available?

Anthropic lists it as active legacy. That means it is an older model, not an announced shutdown. App access is a separate question, so check EZ Ai Assist for the models currently offered.

Did the September price increase happen?

No. Anthropic says the proposed increase to $3 input/$15 output per million did not occur. The $2/$10 rates are now the Standard rates, not an expired promotion.

Can I turn thinking off?

Sonnet 5 supports the disabled thinking setting. Its default is adaptive thinking with high effort. Sonnet 5.5 has different rules, so do not transfer the disabled setting to that model.

Can I tune temperature or top_p?

The model reference says non-default temperature, top_p, or top_k values are rejected. Use supported effort and prompting controls and check the API guide for valid request parameters.

What should I test before switching to Sonnet 5.5?

Compare answer quality and latency on representative prompts, then check thinking configuration, forced-tool use, thinking-block reuse, and how your UI displays inter-tool progress.

Does the 300K beta change normal output capacity?

No. Normal Messages output remains limited to 128,000 tokens. The 300,000 limit belongs to a separately enabled Message Batches beta, not an ordinary chat request.

Is the minimum retirement commitment a shutdown notice?

No. Not sooner than June 30, 2027 is an availability commitment, not a scheduled retirement. API pricing and lifecycle information also do not define EZ Ai Assist subscription terms.

Check the source

Official documentation

Specifications and API prices checked on . Example prompts and workflow advice are editorial guidance from EZ Ai Assist.

Same provider

More from Anthropic

Anthropic

Claude Sonnet 5.5

Anthropic model for coding and everyday professional work, with adaptive thinking, image input, and a 1M-token context window.

Anthropic

Claude Opus 5.5

Claude Opus 5.5 is Anthropic’s model for demanding coding and analysis, with always-on adaptive thinking and optional API Fast Mode.

Anthropic

Claude Opus 5

Anthropic’s active legacy Opus model for complex coding and analysis, with adaptive thinking and a 1M-token context window.

Anthropic

Claude Fable 5.1

Claude Fable 5.1 is Anthropic’s model for demanding reasoning and long-running agentic work, with always-on adaptive thinking.