⚡️UPDATE: Anthropic just made its top AI model dramatically CHEAPER. Claude Opus 5.5 launched at $4 per million input tokens and $20 per million output tokens. At default setting, Claude Opus 5.5 is now: - 40% CHEAPER than Opus 5 - 60% CHEAPER than Fable 5.1 Anthropic says Opus 5.5 matches Fable 5.1 on MOST tasks while leading its own lineup on coding and knowledge-work benchmarks. GPT-6 Astra still LEADS on AutomationBench and Terminal-Bench-Science 0.1. Frontier AI pricing is falling fast even as model performance keeps climbing, raising pressure on OpenAI, Google and the rest of the market to deliver more capability per dollar.