Claude Fable 5 Scores 95% on SWE-bench, Then Hands Off to Opus 4.8
Articles·website·Fable-5 evidence 80%
Anthropic's new Mythos-class model leads on coding benchmarks but deliberately defers to a safer predecessor in restricted domains. That design choice says more than the score.
Signals
created 2026-06-12by peremptory
Fable-5 attribution
dev.to article: Explicit model id: “Claude Fable” (80% confidence)