Ask HN: Did Anthropic Nerf Opus 4.8?
Opus 4.8 was one-shotting simpler bugs just a few days ago for me. The last couple of days however its been like using a slot machine, and I can no longer get it to output clean code that is less-complicated and actually resolves issues. Anyone else seeing this? Opus 4.8 completely misunderstood my request, and took it the opposite way An LLM is not a human and in particular doesn't perform at a consistent level of ability like a human would. So past performance is only a weak indicator of future performance, even without changes to the underlying model. It was always a slot machine.