Releases

Claude Fable 5.2

MiniRouter and Claude Fable 5.2 by Anthropic, separated by a silver fold of light on black.

Anthropic's most capable generally available model now lists 20% below Fable 5.1 and generates up to 30% faster. It leads most of Anthropic's launch table, with large gains on coding, scientific research and remote freelance work.

Claude Fable 5.2 is available on MiniRouter as . The model page has current pricing, provider routes and API setup.

View Claude Fable 5.2

About the model

  • Context: 1M tokens. 128K maximum output.
  • Thinking: Adaptive, always on. Effort runs from low to max; the API default is high.
  • Speed: Up to 30% faster. Output generation against Fable 5.1, per Anthropic.
  • Knowledge cutoff: June 2026. Reliable and training cutoffs per the model docs.

Publisher pricing

Anthropic's list prices, before MiniRouter's fee. Cache reads are $0.20 and five-minute cache writes $10 per million tokens. Fable 5.1 listed $10 input and $50 output.

Publisher-listed API price phases in US dollars per million tokens.
PhaseInput / 1MOutput / 1MBlended / 1MEffectiveSource
List$8.00$40.00$16.00Nov 9, 2026 – open-endedAnthropic ↗

See current MiniRouter rates and how billing works.

Publisher-reported benchmarks

Results from the publisher’s announcement, grouped by unit.

BenchmarkReported resultReported byAs ofSource
FrontierCode 2.049.1%AnthropicNov 9, 2026Announcement ↗
FrontierSWE v269.9%AnthropicNov 9, 2026Announcement ↗
Terminal-Bench-Science 0.174.6%AnthropicNov 9, 2026Announcement ↗
AutomationBench51.4%AnthropicNov 9, 2026Announcement ↗
Remote Labor Index41.3%AnthropicNov 9, 2026Announcement ↗
Humanity's Last Exam · with tools74.2%AnthropicNov 9, 2026Announcement ↗

Fable 5.2 does not lead every row: Anthropic's table puts Gemini 4 Carbon ahead on AutomationBench and Claude Opus 5.5 ahead on the AA-Briefcase Elo rating.

Anthropic's API rejects sampling parameters, disabled thinking and forced tool use. Through MiniRouter those requests succeed, but the settings are not applied.

Anthropic says Fable 5.2 and Mythos 5.2 are the same underlying model; Mythos 5.2 is limited to vetted organizations. Fable 5.2 requires 30-day data retention.

Keep exploring

Compare published results in the model rankings and coding benchmarks, or read more model releases.

For MiniRouter product updates, visit the changelog.

Sources

  1. Anthropic Claude Fable 5.2 announcement

    Published Nov 9, 2026

    Read at Anthropic ↗
  2. Claude Fable 5.2 model documentation

    Published Nov 9, 2026

    Read at Anthropic ↗
  3. What's new in Claude Fable 5.2

    Published Nov 9, 2026

    Read at Anthropic ↗
  4. Claude Fable 5.2 migration guide

    Published Nov 9, 2026

    Read at Anthropic ↗
  5. Vercel Claude Fable 5.2 changelog

    Published Nov 9, 2026

    Read at Vercel ↗

Editorial reference revised 2026-11-09. Catalog status comes from the generated routing snapshot.

Every model, one key.