The latest in AI, every dayAI News
← AI News

November 10, 2026 · Anthropic

Anthropic Launches Claude Fable 5.2 and Claude Mythos 5.2 for the Toughest Technical Work

My take: Anthropic just launched two new Claude models, Fable 5.2 and Mythos 5.2, built for the most ambitious technical work: advanced coding, science, and financial research. What catches my attention isn't just that they're faster (up to 30% faster than the previous version), but what people are already using them for.

Quant finance firms like Jump Trading and XTX Markets report real improvements in how they automate their research. A legal AI company says the model gave better practical guidance on a compliance review. And Oracle found that Fable 5.2 generated correct SQL code where other models got the calculations wrong. This isn't science fiction, it's people using the tool to work better, today.

What I like most is that Anthropic isn't saying "use the biggest model for everything." It recommends Opus 5.5 for everyday work, and saves Fable and Mythos for the truly hard tasks. That's the lesson I keep repeating: AI makes you more efficient when you know how to pick the right tool for each task, not when you default to the most powerful one out of habit.

If you work in something technical (code, financial analysis, research), have you tried delegating the hardest tasks to AI and leaving the simple ones for lighter tools?

Read at the source: Anthropic ↗

Want to use these tools? See the unbiased reviews or back to the news.