Anthropic: Claude Opus 4.6

Claude Opus 4.6 is a text generation model from Anthropic, released on 4 February 2026. Anthropic describes it as their strongest model for coding and long-running professional tasks, built for agents that operate across entire workflows rather than single prompts. It sits in the text category and is aimed at developers who need a capable general purpose model for demanding, multi-step work.

The model accepts text, images and files as input, and returns text as output. It has a context window of 1,000,000 tokens and a maximum output of 128,000 tokens, so it can hold very large inputs and still produce long responses within the same request. Because images and files are accepted alongside text, it can work with mixed content in a single call rather than requiring separate handling for each input type.

The combination of a very large context window, file and image input, and long output makes Claude Opus 4.6 a reasonable fit for tasks that involve reading and reasoning over large amounts of material. Examples include working through an entire codebase, processing long document sets, and supporting agent style workflows where a model needs to track state across many steps. Its support for tool calling and structured outputs also lets it act as the reasoning component in a pipeline that needs to invoke external functions or return responses in a defined format.

You can call it through api14's OpenAI-compatible API using the model id "anthropic/claude-opus-4.6". Requests support the following parameters: include_reasoning, max_completion_tokens, max_tokens, reasoning, reasoning_effort, response_format, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_p and verbosity. Reasoning related parameters such as reasoning, reasoning_effort and include_reasoning let you control how much internal reasoning the model performs and whether it is returned, while tools and tool_choice cover function calling and response_format and structured_outputs cover constrained output shapes.

Since the model returns only text, workloads that need image, audio or video generation are outside its scope and would require a different model. The supported parameters listed above are the ones to rely on when configuring a request; anything not listed should not be assumed to work. For simpler or shorter tasks, a smaller model may be sufficient, but where long inputs, long outputs and agent style behaviour matter, Claude Opus 4.6 is the option Anthropic positions for that work.