name = "Claude Fable 5"
description = "Claude model for creative writing, analysis, and controlled agent workflows"
family = "claude-fable"
release_date = "2026-06-09"
last_updated = "2026-06-09"
attachment = true
reasoning = true
temperature = false
tool_call = true
open_weights = false
knowledge = "2026-01-31"

[limit]
context = 1_000_000
output = 128_000

[modalities]
input = ["text", "image", "pdf"]
output = ["text"]

[[benchmarks]]
name = "SWE-Bench Pro"
score = 80.3
metric = "resolve rate"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"

[[benchmarks]]
name = "SWE-Bench Verified"
score = 95
metric = "resolved"
source = "https://benchlm.ai/benchmarks/sweVerified"

[[benchmarks]]
name = "Terminal-Bench"
score = 88.0
metric = "success rate"
version = "2.1"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"

[[benchmarks]]
name = "Humanity's Last Exam"
score = 59
metric = "accuracy"
variant = "no tools"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"

[[benchmarks]]
name = "Humanity's Last Exam"
score = 64.5
metric = "accuracy"
variant = "with tools"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"

[[benchmarks]]
name = "OSWorld-Verified"
score = 85
metric = "success rate"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"

[[benchmarks]]
name = "FrontierCode"
score = 29.3
metric = "pass rate"
variant = "high effort"
dataset = "Diamond"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"

[[benchmarks]]
name = "GDPval-AA"
score = 1932
metric = "Elo"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"

[[benchmarks]]
name = "AutomationBench"
score = 17.4
metric = "success rate"
source = "https://www.anthropic.com/news/claude-fable-5-mythos-5"
date = "2026-06-09"