Anthropic: Claude Sonnet 4.6 (batch)

Claude Sonnet 4.6 (batch) is a text generation model from Anthropic, offered through api14 under the model id anthropic/claude-sonnet-4.6:batch. It belongs to Anthropic's Sonnet class, which the provider describes as its most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work. The batch label marks this as the batch variant of the model on api14.

The model accepts text, images, and files as input, and it produces text as output. Its context window is 1,000,000 tokens, and the maximum output length is 128,000 tokens. That combination of a very large context and a large output ceiling means you can supply long documents, whole codebases, or extended conversation histories in a single request and still receive a substantial response.

Those capabilities point to a fairly specific set of use cases. Iterative development and complex codebase navigation are areas Anthropic highlights, so tasks such as reviewing large repositories, tracing how a change propagates through many files, or working through multi step coding problems fit well. The same context size suits long document analysis, where an image or file input can be combined with a lengthy text prompt. Agent style workflows also follow naturally, since the model supports tool use and can carry a long running task across many turns. End to end project management and other professional work, as described by the provider, are further cases where a million token context and a large output budget are useful.

You call the model through api14's OpenAI compatible API using the model id anthropic/claude-sonnet-4.6:batch, with the same request shape you would use for other chat models on the gateway. The supported parameters are include_reasoning, max_tokens, reasoning, reasoning_effort, response_format, stop, structured_outputs, temperature, tool_choice, tools, top_p, and verbosity. Tool calling is configured through tools and tool_choice, structured results through response_format and structured_outputs, and reasoning behaviour through reasoning, reasoning_effort, and include_reasoning. Sampling is controlled with temperature and top_p, and verbosity adjusts response length alongside max_tokens.

Because this is the batch variant, it is worth checking whether batch processing matches how you intend to use the model, since interactive or latency sensitive applications may be better served by a non batch version of the same model. Beyond that, the facts here do not indicate any task the model cannot handle, so the choice mostly comes down to whether you need Anthropic's Sonnet class capabilities at this context size or a different model family entirely.