Unofficial status page ·

Claude Fable 5.1

Availability on Claude subscription plans, the new cache-read pricing, and independent benchmark standing. News and system status are monitored continuously by Firecrawl /monitor.

The number five composed of several butterflies

Included in subscription Standard Fable 5.1 is included as a standard part of Max plans, premium seats on Team plans, and premium seats on seat-based Enterprise plans. You can put up to 50% of your weekly usage limits on Fable models at no extra cost; past that, you continue on usage credits or switch models.

Claude system status Operational No known incidents affecting Fable 5.1, per status.claude.com.

Pro & Team Standard Usage credits Fable 5.1 is not included in Pro or Team Standard usage limits - it runs on pay-as-you-go usage credits from the start. The one-time $100 credit from July applied to the Fable 5 change only; there is no equivalent credit for 5.1.

What changed in 5.1

Cache reads: $0.25 / M75% cheaper than Fable 5. Input and output prices are unchanged at $10 / $50 per million.
Typical workload cost: ~25% lessAnthropic's indexed measure vs Fable 5. Up to ~45% less on context-heavy, tool-heavy agentic work.
Default effort: High / MediumHigh in Claude Code; Medium in Claude Cowork and on Claude.ai.
Model ID: claude-fable-5-1Available on the Claude API, AWS, Google Cloud, and Microsoft Azure.

Safeguards now fire less often on benign work.

Claude Code users can expect around 60% fewer interventions per session from the cyber safeguards versus Fable 5, and Fable 5.1 may now be used to identify software vulnerabilities (though not to develop exploits). Biology safeguards fire 85% less often on elementary biology and medical questions. Penetration testing, exploit generation, binary vulnerability scanning, and life-sciences R&D are still redirected to the Opus models. Mythos 5.1 - the same model with more permissive safeguards - stays limited to the Cyber Verification and Life Sciences Verification programs.

Where it lands on benchmarks

Fable 5.1 at max effort is #1 on the Artificial Analysis Intelligence Index with a score of 66, ahead of Opus 5 (63) and Fable 5 (62). It is also slower than average and the most verbose model on the index.

See the full benchmark breakdown →

Fable 5.1 news

Live updates powered by Firecrawl /monitor - new items appear automatically when meaningful Fable news is detected.

DateSourceTitle

Sep 2, 2026 anthropic.com
Claude Fable 5.1 and Mythos 5.1 launch
Available today on all platforms plus AWS, Google Cloud, and Microsoft Azure as claude-fable-5-1. Anthropic reports 52.6% on Terminal-Bench-Science 0.1 (vs 24.7% for Fable 5), 55.8% on Terminal-Bench 4.0, and 73.4% on CursorBench 3.2.0. Fable 5.1 and Mythos 5.1 are the same underlying model with different safeguards.
Sep 2, 2026 anthropic.com
Cache reads drop 75%, cutting typical costs by a quarter
Cache reads now cost $0.25 per million tokens. Input and output pricing is unchanged at $10 / $50 per million, but because cache reads dominate agentic workloads, Anthropic measures around 25% lower cost for typical workloads and up to about 45% for context-heavy agentic work.
Sep 2, 2026 support.claude.com
No new free credit for Pro and Team Standard
Fable 5.1 inherits Fable 5's plan terms: included at up to 50% of weekly limits on Max and premium Team/Enterprise seats, usage credits on Pro and Team Standard. Anthropic's help centre confirms the one-time credit from July covered the Fable 5 change only, with no equivalent for 5.1.
Sep 2, 2026 artificialanalysis.ai
Artificial Analysis puts Fable 5.1 at #1 on intelligence
An Intelligence Index score of 66 at max effort, ahead of Opus 5 at 63 and Fable 5 at 62, at $3.69 per index task. Fable 5.1 also produced 140M output tokens across the index against a 71M median, making it the most verbose model measured.
Sep 2, 2026 anthropic.com
Enterprise Frontier Safeguards bring zero-retention privacy
EFS stores customer data in the customer's own cloud infrastructure rather than Anthropic's, giving enterprises the privacy of a zero-data-retention agreement while keeping misuse detection. It rolls out in phases starting this fall; eligible customers can use Fable 5.1 with zero data retention until then.
Sep 2, 2026 anthropic.com
Anti-distillation change affects new API accounts
API accounts created from today onwards can no longer manually edit Claude's prior context in a multi-turn conversation while preserving the transcript of its prior thinking, closing a documented distillation technique. Existing accounts are unaffected for now but the change applies to all users with future releases.
Jul 24, 2026 anthropic.com
Claude Opus 5 launches at half Fable 5's price
Opus 5 is available on all platforms at $5 / $25 per million input/output tokens - half Fable's $10 / $50 - and becomes the new default model on Claude Max. Blocked biology requests on Fable now route to Opus 5.
Jul 18, 2026 x.com/claudeai
Fable becomes standard on Max and Team Premium
From July 20, Fable is included at 50% of limits on all Max and Team Premium plans - no longer a temporary extension. Pro and Team Standard keep access via usage credits and received a one-time $100 credit.
Jun 9, 2026 anthropic.com
Claude Fable 5 and Mythos 5 launch
A Mythos-class model made safe for general use: state-of-the-art on nearly all tested benchmarks, 1M-token context, priced at $10 / $50 per million input/output tokens.