RIP GPT 6: OPUS 5.5 JUST DESTROYED OPENAI (INSANE)

Anthropic just dropped Claude Opus 5.5, and it's nuts. It performs at the level of Claude Fable 5.1 on most work, it's the new top model on agentic coding, and it costs 40% less to run than Opus 5. I put it straight into Claude Code with my standard website prompt, and the result is the best-looking build I've had from any model. A full luxury Rolls-Royce experience site, cream and gold, illustrated Vesuvius coastline, route timeline, the lot. The numbers (from Anthropic's announcement): • Terminal-Bench 4.0: Opus 5.5 66.4% vs GPT-6 Astra 57.9%, Fable 5.1 55.8%, Opus 5 52.3% • Output: $20 pe