Anthropic launches Opus 5: stronger, cheaper, and less restricted

Friends, I’d like to share some news from the AI world: Anthropic has released Opus 5.

The model is now cheaper and less restricted than Fable 5, while delivering better results on several benchmarks.

Anthropic also highlights that Opus 5 is better at checking its own work and more reliably carries tasks through to completion.

For API users, the updated safety mechanisms matter too: when filters are triggered, Automatic Fallbacks now route the request to a simpler model instead of returning an error.

Why this matters: the AI market is once again shifting toward practicality — higher quality, fewer limitations, and clearer use cases.

Have you already tested Anthropic’s new models?