Anthropic: Claude Sonnet 5.5

Claude Sonnet 5.5 is a text model from Anthropic, released on 28 September 2026 as a direct upgrade to Claude Sonnet 5. It sits in Anthropic's Sonnet class, which the provider describes as aimed at well-scoped everyday work rather than the heaviest frontier tasks. On api14 it is listed under the model id anthropic/claude-sonnet-5.5 and its main category is text.

The model accepts text, images and files as input, and it returns text only. Its context window is 1,000,000 tokens, which is large enough to hold entire codebases, long document sets or extended conversation histories in a single request. The maximum output is 128,000 tokens, so responses can run to substantial length when a task calls for it. Anthropic highlights feature building and bug fixing as particular strengths, alongside producing written output.

Those capabilities point to a fairly specific set of uses. The combination of a very large context window and image and file input suits code assistance on real repositories, where a developer wants the model to read many files at once and propose changes or trace a bug across them. The same input mix works for document analysis, such as summarising or extracting structured data from long reports, PDFs or scanned material. Because output can reach 128,000 tokens, the model can also handle generation tasks that need long, continuous responses rather than short answers. Tool calling and structured outputs make it a reasonable fit for agent style workflows and for pipelines that need machine readable results.

You call it through api14's OpenAI-compatible API using the model id anthropic/claude-sonnet-5.5. The supported request parameters are include_reasoning, max_completion_tokens, max_tokens, reasoning, reasoning_effort, response_format, stop, structured_outputs, tool_choice, tools and verbosity. In practice that means you can cap output length, control how much reasoning effort the model applies and whether reasoning is surfaced, constrain responses to a schema through response_format or structured_outputs, define tools and how the model may choose among them, set stop sequences, and adjust verbosity. Code written against the OpenAI chat completions shape should work with only the base URL and model id changed.

The main thing to weigh is scope. Anthropic positions this as a Sonnet-class model for well-scoped everyday work, so if your task is at the extreme end of difficulty or needs output beyond text, another model may fit better. Output is text only, so image or audio generation is out of scope here. Within text work, though, the million token context and 128,000 token output ceiling cover a wide range of development and document tasks.