Skip to content

AI Agents & Coding · Paid

Claude Fable

Claude Fable is Anthropic's top tier of Claude models, positioned above Opus, Sonnet and Haiku for the most ambitious, open-ended technical work: long multi-step software projects, large migrations, research and analysis over big document sets, and coordinating teams of other agents. The current version, Claude Fable 5.2, was released on November 9, 2026, replacing Fable 5.1 as the newest model in the line. Anthropic describes Fable and Claude Mythos as the same underlying model: Fable ships with strong cybersecurity and biology safeguards so it can be offered to all users, while Mythos is reserved for organizations vetted through Anthropic's Cyber and Life Sciences Verification Programs.

Who it is for: engineering teams and technical founders with hard, long-running tasks where model quality matters more than per-token cost, such as agentic coding on unfamiliar codebases, multi-agent orchestration, hardware and scientific workflows, and deep research. Anthropic itself recommends starting most coding and knowledge work on Claude Opus 5.5 and moving up to Fable only when the scope or difficulty calls for it. Fable can also act as an advisor that smaller models consult, or as an orchestrator that delegates simpler steps to them.

Key features of Fable 5.2: - A 1 million token context window and up to 128,000 output tokens, with long context billed at standard rates. - Adjustable effort levels (from Low up to Max), so teams can trade quality against tokens and cost per task. - An optional fast mode that Anthropic says generates tokens up to 2.5 times faster, at premium rates. - Prompt caching, with cache hits billed at 2.5% of the base input price, and the Batch API at half price for asynchronous work. - Availability on the Claude Platform (model ID claude-fable-5-2), in Claude apps and Claude Code on paid plans, and through Amazon Web Services, Google Cloud and Microsoft Azure. - Safety classifiers that block high-risk cybersecurity tasks and most research biology tasks, answering those requests with a fallback model instead.

Pricing type: paid. As of November 10, 2026, Anthropic lists Claude Fable 5.2 at $8 per million input tokens and $40 per million output tokens on its API, $16 and $80 in fast mode, and $4 and $20 through the Batch API. Partner clouds such as Amazon Bedrock and Google Cloud set their own prices. Paid Claude subscriptions include Fable usage within plan limits on higher tiers; check Anthropic's pricing and help center pages for current terms.

Alternatives: Claude Opus 5.5 (Anthropic's recommended daily driver at half the per-token price), OpenAI's GPT-6 models, Google's Gemini 4 models, and open-weight models such as Kimi and Qwen for teams that need self-hosting or lower costs.

Official URL: https://www.anthropic.com/claude-fable-and-mythos-5-2

Official site