Anthropic releases Opus 5 with ‘close’ to Fable 5’s capabilities

Anthropic has released its new AI model, Opus 5, which features enhanced cybersecurity safeguards due to government concerns and is comparable to the capabilities of the previous Fable 5 model, particularly in complex coding tasks. The company emphasizes that Opus 5 has stronger cybersecurity measures than its predecessor and is more resistant to misuse, while pricing remains consistent with Opus 4.8 and is cheaper than Fable 5 and GPT-5.6.

• Anthropic released its newest model, Opus 5, with enhanced cybersecurity safeguards.
• Opus 5’s capabilities are described as close to those of Fable 5, especially in complex coding tasks.
• The model has undergone independent testing by government partners.
• Pricing for Opus 5 is the same as its predecessor and cheaper than Fable 5 and OpenAI’s GPT-5.6.
• New features include a ‘Fast mode’ for increased speed and an option for automatic fallbacks to lower-tier models when safeguards are triggered.
• Opus 5 is marketed towards enterprise customers for knowledge work and biology, while Fable 5 remains for more ambitious projects.
  Continue reading https://www.theverge.com/ai-artificial-intelligence/970105/claude-opus-5-announced-anthropic-ai-model-release