Skip to content
Book a callSkip the hiring grind

Claude Fable 5.1

Anthropic · Released 1 Sep 2026 · claude-fable-5-1

Compare this model
ProprietaryReasoningAgentsCodingResearchLong context

$10.00

Input price / 1M tokens

#59 of 61, cheapest first

$50.00

Output price / 1M tokens

#59 of 61, cheapest first

1M

Context window

#29 of 71

68

Output tokens / second

#46 of 61

55.8%

Terminal-Bench 4.0

#5 of 10

77.9%

OSWorld 2.0

#1 of 10

60.9%

Humanity’s Last Exam (no tools)

#1 of 23

65%

Humanity’s Last Exam (with tools)

#2 of 16

Summary

Claude Fable 5.1 is a proprietary model from Anthropic, released on 1 Sep 2026.

At $10.00 input and $50.00 output per million tokens, it is #59 of 61 on input price, cheapest first.

Its 1M-token context window ranks #29 of 71.

Measured output speed is 68 tokens per second, #46 of 61.

Anthropic's top generally available model, for demanding reasoning and long-horizon agentic work. Successor to Claude Fable 5 at the same input and output prices, with stronger long-running agentic coding, multistep research and document work.

Benchmarks · 14 reported

BenchmarkScoreBarRankSource
Terminal-Bench 4.0agentic coding55.8%#5 of 10Anthropic Fable 5.1 / Mythos 5.1 launch postSelf-reported
Terminal-Bench-Science 0.1agentic scientific research; SE ±3.5–4.5 pts52.6%–Anthropic Fable 5.1 / Mythos 5.1 launch postSelf-reported
GDPval-AA v2knowledge work1,853Elo-style–Anthropic Fable 5.1 / Mythos 5.1 launch postSelf-reported
OSWorld 2.0partial-credit score; benchmark authors' August 2026 task release77.9%#1 of 10Anthropic Fable 5.1 / Mythos 5.1 launch postSelf-reported
OSWorld 2.0strict score; benchmark authors' August 2026 task release41.7%–Anthropic Fable 5.1 / Mythos 5.1 launch postSelf-reported
Humanity’s Last Exam (no tools)no tools60.9%#1 of 23Anthropic Fable 5.1 / Mythos 5.1 launch postSelf-reported
Humanity’s Last Exam (with tools)with tools65%#2 of 16Anthropic Fable 5.1 / Mythos 5.1 launch postSelf-reported
AutomationBenchbusiness workflows (Zapier benchmark)31.4%#5 of 8Anthropic Fable 5.1 / Mythos 5.1 launch postSelf-reported
CursorBench 3.2.0agentic coding73.4%–Anthropic Fable 5.1 / Mythos 5.1 launch postSelf-reported
FrontierCode 1.1agentic coding; comparison column in a later post50.3%#2 of 6Anthropic Opus 5.5 launch postSelf-reported
CursorBench 4.0agentic coding; comparison column in a later post51.8%–Anthropic Opus 5.5 launch postSelf-reported
GDPval-AA v2.1knowledge work; comparison column in a later post1,735Elo-style–Anthropic Opus 5.5 launch postSelf-reported
OSWorld 2.1partial-credit score; comparison column in a later post80.7%–Anthropic Opus 5.5 launch postSelf-reported
Chartographywith tools; comparison column in a later post88.4%–Anthropic Opus 5.5 launch postSelf-reported

Self-reported means the lab ran the test itself. A rank appears only where several labs report the same version of a benchmark.

Similar models

Models with the most benchmarks in common with Claude Fable 5.1, and the closest scores.

Compare 4 side by side

Terminal-Bench 4.0

#5 of 10

Long jobs in a real terminal

OSWorld 2.0

#1 of 10

Long desktop and web workflows

Humanity’s Last Exam (no tools)

#1 of 23

Expert questions, no search or code

Humanity’s Last Exam (with tools)

#2 of 16

Expert questions, with search and code

Compare Claude Fable 5.1 side by side

Pre-filled with the two closest models. Swap any of them.

Open comparison

Sources

Building on Claude Fable 5.1?

Get the architecture right before the bill arrives.

We size caching, routing and fallbacks for your workload on a free call, and tell you where a cheaper model is good enough.

Book a free call All AI models