Head-to-head token pricing, context window limits, and blended inference cost comparison between Claude Sonnet 5.5 (Anthropic) and Claude Opus 5.5 (Anthropic).
Claude Opus 5.5 costs 2.0x more than Claude Sonnet 5.5 on a 3:1 blended token workload. Claude Sonnet 5.5 has a context window of 1000k tokens, while Claude Opus 5.5 provides 1000k tokens.