Anthropic Drops Claude Fable 5.2 and Mythos 5.2 — and Cuts Prices 20%

Anthropic just released its two most powerful models, Claude Fable 5.2 and Claude Mythos 5.2, and it did something unusual at the same time: it lowered the price.

The two models went live on November 9. Fable 5.2 is the new general-purpose flagship, aimed at the hardest technical work. Mythos 5.2 runs the same underlying model but is reserved for cybersecurity and life sciences organizations that pass Anthropic’s verification process.

The headline number for most engineering teams is the price cut. Standard API rates drop to $8 per million input tokens and $40 per million output tokens, down 20% from the $10 and $50 that Fable 5.1 charged. Anthropic also says the new models generate tokens up to 30% faster than their predecessors, with an optional Fast mode that can push speed up to 2.5 times at double the standard rate.

The benchmark results put real distance between Fable and the rest of the field. On FrontierCode 2.0, which asks whether a patch is good enough to merge into a real open-source repository, Fable 5.2 scored 49.1%, ahead of OpenAI’s GPT-6 Astra at 42.8% and Google’s Gemini 4 Carbon at 41.1%. On FrontierSWE v2, the margin was similar: 69.9% for Fable, 65.5% for Astra, 61.2% for Carbon.

But the number that matters more than any benchmark may be what a task actually costs. Anthropic’s system card puts the average cost of a FrontierCode 2.0 task at $6.72 for Fable 5.2 on its High setting — nearly identical to what GPT-6 Astra spends at maximum effort ($6.68), and roughly half of what Opus 5.5 spent at its max ($12.82) for a nearly identical score.

The timing is not a coincidence. The launch lands in the middle of Anthropic’s IPO roadshow, with media reports pointing to a Nasdaq debut around November 19 or 20. The company has not confirmed a date, and valuation figures remain unreported in official filings.

The takeaway for buyers: model competition has quietly moved from “which model is strongest” to “which model finishes a hard job for less.” On that second question, Anthropic just made its clearest argument yet.