SAN FRANCISCO, November 9 — Anthropic released Claude Fable 5.2 on Monday with lower API prices and output generation up to 30 percent faster than Fable 5.1, the company said.
Standard rates are $8 per million input tokens and $40 per million output tokens. Computerworld reported a 20 percent reduction from $10 for input and $50 for output. Anthropic said optional fast mode costs $16 per million input tokens and $80 per million output tokens, with generation up to 2.5 times faster.
Anthropic said Fable 5.2 shares its underlying model with Mythos 5.2, but access and safeguards differ. Its announcement also addressed the Life Sciences program and Cyber Verification program.
On the FrontierCode 2.0 coding evaluation, Anthropic reported Fable 5.2 scored 49.1 percent at Xhigh effort, compared with 32.4 percent for Fable 5.1.
At High effort, Fable 5.2 scored 45.3 percent against 45.5 percent for Opus 5.5 at Max effort, at roughly half the cost per task, according to Anthropic’s evaluation. Medium, rather than High or Xhigh, is the default effort setting in Claude apps and Claude Code, the company said.
“Anthropic has cut the price of a token, not the price of a thought,” Greyhound Research chief analyst Sanchit Vir Gogia said, according to Computerworld, urging measurement against successfully completed tasks.
For small B2B businesses whose products or sales processes depend on software, the reported rates and speed suggest potentially cheaper or quicker testing of customer-facing improvements and lead-generation workflows. That is an inference, not evidence of better conversions or customer growth. Task-specific comparisons would need to assess cost per successful result, rather than token prices alone.
Anthropic advised starting with Opus 5.5 for most coding and knowledge tasks, reserving Fable for more ambitious, open-ended technical work.