Anthropic Super-Frontier Available in FlowHunt

Claude Fable 5.1

Claude Fable 5.1 is a Super-Frontier AI model developed by Anthropic and released in September 2026. It is a proprietary closed-weight model offered for general use through Anthropic. The 1M-token context window is large enough to hold entire codebases, lengthy technical documents, or extended agentic sessions without truncation.

Read more
Claude Fable 5.1 and Claude Mythos 5.1 are two configurations of the same model, sharing identical weights. Fable 5.1 is the generally available version and adds safeguards that block tasks in high-risk, dual-use domains such as biology and cybersecurity. Mythos 5.1 relaxes those safeguards for vetted individuals and organizations through Anthropic’s trusted access programs. Anthropic says the model advances coding, knowledge work and problem-solving, with better novel mathematical and scientific reasoning, and that Fable 5.1’s safeguards produce fewer false positives than Fable 5 had at launch. It is only available with thinking enabled.
Try in FlowHunt Back to AI Leaderboard

Benchmarks

Each score appears once, with its rank among tracked models and who measured it.

Compared with Super-Frontier peers on the same scale
  • Claude Fable 5.1this model
  • GPT-6 Astrathis model leads on 2 of 6
  • Claude Fable 5this model leads on 6 of 7
Verified independent evaluator, counted in the ranking Self-reported published by Anthropic, shown but not ranked
Overall

AA Intelligence Index

Independent composite of agent, coding, knowledge and science tests by Artificial Analysis (v4).

Source ↗

Claude Fable 5.1 53.4 GPT-6 Astra 52.7 Claude Fable 5 49.6
#3of 45 Verified
Coding

DeepSWE

113 long-horizon software tasks written from scratch, graded by hidden tests.

Source ↗

Claude Fable 5.1 67.4% GPT-6 Astra 74.1% Claude Fable 5 69.9%
Not ranked Self-reported

SWE-bench Pro

Real issues across several languages and large codebases.

Source ↗

Claude Fable 5.1 81.2% GPT-6 Astra No score Claude Fable 5 80.3%
Not ranked Self-reported
Agents

Terminal-Bench 4.0

Multi-step professional tasks done by an agent in a command-line terminal.

Source ↗Also: Self-reported 55.8% (Fable 5.1 per Opus 5.5 launch table (system card rounds to 56%; Mythos 5.1, same model with fewer safeguards, 60.9))

Claude Fable 5.1 52.0% GPT-6 Astra 59.1% Claude Fable 5 42.4%
#5of 36 Verified
Reasoning

Humanity's Last Exam

Expert-written questions across 100+ subjects, no tools.

Source ↗Also: Self-reported 60.9% (no tools)

Claude Fable 5.1 59.1% GPT-6 Astra 54.7% Claude Fable 5 55.5%
#2of 44 Verified

GPQA Diamond

Graduate-level science questions that search cannot answer.

Source ↗Also: Verified 93.4% (Vals AI own implementation)

Claude Fable 5.1 93.7% GPT-6 Astra 96.1% Claude Fable 5 92.6%
#5of 37 Verified

ARC-AGI-2

Novel visual puzzles that cannot be memorised.

Source ↗

Claude Fable 5.1 90.0% GPT-6 Astra 95.0% Claude Fable 5 89.2%
#6of 27 Verified

What to Use Claude Fable 5.1 For

Software Engineering

Anthropic positions it as an advance in coding, so it fits complex, multi-step engineering work and agentic coding workflows where you want the strongest Claude model on the task.

Scientific & Mathematical Reasoning

Use it for problems that need novel mathematical or scientific reasoning, such as research questions, technical analysis and multi-discipline work.

Knowledge Work

It handles professional knowledge work such as analysing long documents, drafting reports and reasoning across large bodies of source material.

Large Codebase & Document Analysis

With a 1M-token context window, Claude Fable 5.1 can take in entire repositories, long contracts or extended working sessions in one prompt, without chunking or a retrieval pipeline.

Strengths and limitations

Strengths

  • Strong coding and agentic work
  • Novel mathematical and scientific reasoning
  • Knowledge work across long documents
  • Very long documents and large codebases (1M context)
  • Updated safeguards with fewer false positives than Fable 5 at launch

Limitations

  • Closed-weight, so it cannot be self-hosted or fine-tuned on private data
  • Safeguards block some dual-use biology and cybersecurity tasks; unrestricted access is limited to vetted organizations via Mythos 5.1
  • Thinking is always enabled, and Anthropic says Opus 5.5 performs at its level on most work

Use it in FlowHunt

Use Claude Fable 5.1 in any flow, next to your own tools and knowledge sources.

  1. 1Open a flow, or create one from a template
  2. 2Add an AI Agent or Generator component
  3. 3Pick Claude Fable 5.1 in the model selector

Related AI Models

About Anthropic San Francisco, CA · Founded 2021 · anthropic.com

Anthropic is an AI safety company founded in 2021 by former OpenAI researchers Dario Amodei and Daniela Amodei, along with a group of colleagues who had led OpenAI's safety and policy teams. The company's stated mission is the responsible development of AI for the long-term benefit of humanity, and its research agenda is defined by that commitment: Constitutional AI (CAI), a technique for training models to be helpful, harmless, and honest through self-critique; mechanistic interpretability, which aims to reverse-engineer what individual neurons and circuits inside large networks actually compute; and scaling science, which tries to predict how model behaviour changes with compute and data. Anthropic publishes this research openly and has become one of the most cited AI safety labs in academic literature. Claude is Anthropic's public model family, named for Claude Shannon, the founder of information theory, and runs from Claude 1 to the Claude 5.5 generation and Claude Fable 5.1. Anthropic operates a commercial API and consumer-facing Claude.ai product, and maintains deep cloud partnerships with AWS (Amazon Bedrock) and Google Cloud (Vertex AI).

anthropic.com ↗

Browse the Full AI Model Leaderboard

Compare every tracked model on coding, agent and reasoning benchmarks, with each score marked verified or self-reported.

← Back to AI Leaderboard Try in FlowHunt

Frequently asked questions