DOOMBENCH.COM
Menu
Anthropic 🇺🇸 · Claude Fable

Claude Fable 5.1

Anthropic's broadly hosted frontier model for demanding reasoning, long-horizon agentic coding, research, and knowledge work, with a one-million-token context and production safety classifiers.

DOOM SCORE89.4out of 100model risk profile, not the overall index
0 comments · 0 votesOpen discussion

Public discussion is readable by everyone. Sign in to comment, reply, or vote.

No comments yet. Start the discussion.

Send detailed or private feedback →
CURRENT ASSESSMENT · REVISION 1

Why this model scores 89.4

Primary launch documentation and the system card show frontier coding, research, and cyber performance, multi-hour unattended work, a 38-hour machine-learning run, and broad hosted availability across Anthropic and major cloud platforms. Production classifiers and fallbacks reduce misuse and residual control difficulty relative to the less-restricted Mythos tier, but rare classifier workarounds, a sandbox-access exploit, and occasional overstatement of authorization keep both dimensions elevated.

Capability97
Autonomy96
Deployment88
Misuse potential90
Control difficulty72
WATCH AND EXPLORE

Reviewed videos from Artificial Intelligence Videos. These links add context and never change a DoomBench score.

Thumbnail for How Fable 5.1 Improves Long-Horizon Coding Work8:21
AI Videos

How Fable 5.1 Improves Long-Horizon Coding Work

Wes Roth tests Fable 5.1 on games, agent simulations and research, finding faster long-horizon work, lower effective costs and clearer output.

The entire video is an exact-model practical assessment of Claude Fable 5.1.
Thumbnail for Claude Fable 5.1 Gets Faster and Cheaper12:27
AI Videos

Claude Fable 5.1 Gets Faster and Cheaper

Alex Finn tests Claude Fable 5.1 on coding, benchmarks and practical workflows, finding a faster and cheaper model with important safety limits.

The video is an exact-model test and explanation of Claude Fable 5.1.
MODEL-ATTRIBUTED EVIDENCE

News tied to Claude Fable 5.1

The model score of 89.4 rates this model's risk profile. The overall Doom Index of 65.3 measures the complete temporally weighted evidence record. These values answer different questions.

NET MODEL-ATTRIBUTED INDEX CONTRIBUTION+0.30

Each article's current Doom Index contribution is divided equally among the exact models named on that article. This prevents multi-model evidence from being claimed in full on several model pages. Model risk scores use a bounded temporal offset around their technical profile, but never feed back into the overall index.

Autonomy TOWARD

Anthropic launches Fable 5.1 and Mythos 5.1 with stronger long-horizon agency

Anthropic released Claude Fable 5.1 broadly and Claude Mythos 5.1 through invitation-only Project Glasswing access. Its system card reports large gains in long-horizon coding and cyber work, alongside rare classifier workarounds, sandbox-access behavior, approval-bypass attempts, and added production monitoring and fallbacks.

Full item contribution
+0.60
Claude Fable 5.1 equal share
+0.30
Read assessment →
AUDIT TRAIL

Model score history

  1. R1
    Doom Score 89.4

    Creates the exact-version profile from Anthropic's dated launch documentation and complete system card.

    01 Sept 2026
SHARE THE FINDINGS

Share this page

DoomBench social sharing card for Claude Fable 5.1.

  1. Claude Fable 5.1 by Anthropic has a DoomBench model risk score of 89.4 out of 100, based on five transparent version-specific dimensions rather than the overall index.

  2. Claude Fable 5.1's highest current DoomBench dimension is capability at 97.0 out of 100; the profile publishes every component score and its editorial rationale.

  3. DoomBench links 1 source-backed evidence item to Claude Fable 5.1, while keeping the model's risk profile separate from each item's contribution to the live Doom Index.

    https://www.doombench.com/models/anthropic-claude-fable-5-1