Updated
Updated · CNET · Jul 24
Anthropic Releases Opus 5 at Half Fable 5's Price, Adds 2.5x Faster Mode
Updated
Updated · CNET · Jul 24

Anthropic Releases Opus 5 at Half Fable 5's Price, Adds 2.5x Faster Mode

3 articles · Updated · CNET · Jul 24

Summary

  • Anthropic said Opus 5 delivers its strongest work performance yet, pushing Claude closer to Fable 5 capabilities while targeting enterprise use cases such as coding, knowledge work and biology.
  • Half-price positioning is central to the launch: Opus 5 is priced below Fable 5, and a research-preview fast mode generates output tokens 2.5x faster at 2x standard Opus 4 pricing.
  • New autonomy features let the model complete tasks with less user input, double-check its own work and recover from errors, with Anthropic citing notable gains over Opus 4.8 in scientific research, especially organic chemistry.
  • Safety is also a selling point: Anthropic called Opus 5 its most secure model yet, added stricter guardrails and a fallback-model option for blocked prompts, and let developers change tool access midconversation without breaking prompt cache.
  • The release lands in a broader AI model race after Google and OpenAI rolled out newer, more efficient systems earlier this month, sharpening pressure on vendors to improve both performance and cost.

Insights

Why did Anthropic drastically reduce Opus 5's safety triggers, and does this signal a retreat from their strict AI alignment principles?
Can Opus 5 truly distinguish between defensive code scanning and malicious exploit generation without accidentally enabling bad actors?
How will Anthropic’s shift from hard safety blocks to controlled degradation reshape the way enterprise applications handle restricted AI prompts?