What Is Claude Fable 5.1?
As of September 2026, Claude Fable 5.1 is Anthropic’s generally available model for demanding coding, research, document-heavy work, and long-running AI agents. Anthropic positions it for tasks that may require sustained reasoning, multiple tool calls, and work across several applications.
Claude Fable 5.1 was released on September 1, 2026. The Claude Platform documentation lists a 1-million-token context window, a maximum output of 128,000 tokens, always-on adaptive thinking, and a default effort level of high.
Anthropic does not recommend Fable 5.1 for every workload. Its model-selection guidance says to start with Claude Opus 5 for most workloads and consider Fable 5.1 when demanding reasoning, long-horizon agentic work, or internal evaluations justify the additional cost and slower latency.
What Is Claude Mythos 5.1?
Claude Mythos 5.1 uses the same underlying model as Claude Fable 5.1, but Anthropic deploys it under a different access and safety policy. Mythos 5.1 is intended for controlled cybersecurity and biology research rather than general commercial use.
Access is limited to vetted organizations through Anthropic’s trusted access programs. Anthropic says its Life Sciences Verification Program is launching as an invite-only beta, while Mythos access for cybersecurity work is being introduced through its Cyber Verification Program. The company currently says access is limited to a set of U.S. organizations while it works to expand availability.
Claude Fable 5.1 vs. Claude Mythos 5.1
The main difference is therefore not the underlying model. It is the deployment policy, safety layer, and access conditions surrounding that model. Fable 5.1 is designed for broader availability with additional safeguards, while Mythos 5.1 is restricted to approved research environments.
What’s New in Claude Fable 5.1?
Claude Fable 5.1 extends Claude Fable 5 with changes to agentic work, pricing, API controls, and response handling. Anthropic’s documentation identifies five additive changes and three breaking changes for existing Fable 5 integrations.
Stronger long-running agentic work
Anthropic says Fable 5.1 brings stronger long-running agentic coding, multistep research, and work involving documents, spreadsheets, and slides. The product page describes tasks that may take hours, span multiple applications, use tools, recover from failed steps, and provide progress updates.
This makes Fable 5.1 more relevant to workflows that go beyond a single prompt-and-answer exchange, such as autonomous coding sessions, extended research, browser operations, and managed agents.
Lower cache-read pricing
Fable 5.1 keeps the standard price of $10 per million input tokens and $50 per million output tokens, but cache reads now cost $0.25 per million tokens. Anthropic says this is 75% lower than the cache-read price for Fable 5.
Anthropic estimates that the lower cache-read price may reduce costs by around 25% for typical workloads and by up to approximately 45% for highly agentic workloads. These are Anthropic’s estimates, not independent pricing benchmarks.
New API controls and progress updates
The 5.1 release adds several developer-facing features:
- Per-message effort controls, currently in beta
- Turn-scoped system messages, currently in beta
- Readable progress updates between tool calls, currently in beta
- Content provenance features
- Lower cache-read pricing
These changes are especially relevant to developers building long-running agents and applications that use tools repeatedly over an extended workflow.
Breaking changes for Fable 5 integrations
Developers migrating from Claude Fable 5 should review Anthropic’s migration guidance. The documentation identifies three breaking changes:
- Forced tool use now returns an error
- Earlier models cannot read Fable 5.1 thinking blocks
- Editing earlier conversation turns invalidates related thinking blocks
These changes may affect response handling, conversation-state management, and tool-use implementations in production systems.
Coding, Research, and Document Work
Anthropic describes Fable 5.1 as its most capable generally available model for coding and knowledge work. Its listed coding use cases include codebase-wide features, code review, performance work, testing, and multi-day autonomous coding sessions. The model can also use vision to compare outputs with an original design or goal.
For research and enterprise workflows, Anthropic highlights multi-stage analysis and work with diagrams, charts, tables, files, spreadsheets, presentations, and PDFs. These capabilities may be useful in areas such as finance, legal analysis, software architecture, analytics, and other document-heavy workflows.
These descriptions are based on Anthropic’s product materials. They should be treated as vendor claims rather than independent proof that Fable 5.1 is the best-performing model for every task or industry.
Claude Fable 5.1 Pricing
Claude Platform documentation lists the following standard pricing for Claude Fable 5.1:
- Input: $10 per million tokens
- Output: $50 per million tokens
- Five-minute cache writes: $12.50 per million tokens
- One-hour cache writes: $20 per million tokens
- Cache reads: $0.25 per million tokens
- Batch API: 50% discount on standard input and output pricing
For eligible Batch API workloads, the 50% discount lowers the standard rates to $5 per million input tokens and $25 per million output tokens.
Anthropic’s Mythos 5.1 product page lists pricing starting at $10 per million input tokens and $50 per million output tokens. The Claude Platform documentation says Mythos 5.1 shares Fable 5.1’s specifications and pricing, although access remains restricted.
How to Access Claude Fable 5.1
Anthropic lists Claude Fable 5.1 for Pro, Max, Team, and Enterprise users. However, plan-level access and billing treatment can vary, and some users may need usage credits in addition to their regular plan limits.
For developers, Fable 5.1 is available through:
- Claude API
- Amazon Bedrock
- Google Cloud
- Microsoft Foundry
- Claude Platform on AWS
The Claude API model ID is:
claude-fable-5-1
The model is also available through Anthropic’s native Claude Platform and supported cloud marketplaces.
Should You Use Fable 5.1 Instead of Claude Opus 5?
Not necessarily.
Anthropic’s model overview recommends starting with Claude Opus 5 for most workloads. Fable 5.1 is positioned for demanding reasoning and long-horizon agentic work, especially when evaluations show that Opus 5 at higher effort does not meet the required performance level.
Fable 5.1 is therefore better understood as a specialized option for difficult, long-running workloads rather than a universal replacement for Claude Opus 5.
FAQ
Is Claude Fable 5.1 an official Anthropic model?
Yes. Claude Fable 5.1 appears on Anthropic’s official product page and in the Claude Platform documentation. Anthropic lists its release date as September 1, 2026.
Can everyone use Claude Mythos 5.1?
No. Mythos 5.1 is restricted to vetted organizations through Anthropic’s trusted access programs. Anthropic currently says availability is limited to selected U.S. organizations while it works to expand access.
What are the main changes in Fable 5.1?
The main documented changes include stronger long-running agentic work, lower cache-read pricing, per-message effort controls, turn-scoped system messages, readable progress updates between tool calls, content provenance, and several compatibility changes for Fable 5 integrations.
How much does Claude Fable 5.1 cost?
The standard API price is $10 per million input tokens and $50 per million output tokens. Cache reads cost $0.25 per million tokens, and eligible Batch API workloads receive a 50% discount on standard input and output pricing.
Conclusion
Claude Fable 5.1 is not simply a faster or cheaper replacement for every Claude model. It is Anthropic’s generally available option for work that demands sustained reasoning, extended coding, multistep research, tool use, and document-heavy execution.
Its most important practical changes are the lower cache-read price and the new controls for long-running agentic workflows. At the same time, its higher standard price and slower comparative latency mean that Claude Opus 5 may remain the better starting point for many applications.
Claude Mythos 5.1 shows how Anthropic separates model capability from deployment policy. It shares the same underlying model as Fable 5.1, but access and safeguards differ because Mythos is intended for controlled cybersecurity and biology research.
For most developers and businesses, Claude Fable 5.1 is the relevant model to evaluate. Claude Mythos 5.1 remains a restricted offering for approved organizations with specific research and security needs.
