Anthropic: Claude Sonnet 5.5

Claude Sonnet 5.5 is a paid Anthropic model that accepts text, image, and file inputs. It offers a 1,000,000-token context window and up to 128,000 completion tokens per response, and it supports tool use and reasoning. Structured output support is not listed in the available data, so teams that depend on it should confirm before committing. Pricing is $2 per million input tokens and $10 per million output tokens. Output costs five times as much as input, so workloads that generate long responses will drive most of the spend. The model posts a blended benchmark score of 97.0, but that figure rests on a single independent benchmark (an AI Index score of 92.4), so its standing is thinly evidenced. Shortlist it if you need very long context, multimodal file handling, and tool use in one model. Run your own evaluations before treating the benchmark result as settled.