Anthropic reports 64.9% for Claude Fable 5.2 on RoboDojo Sim at Max effort. Anthropic says it ran RoboDojo Sim internally for each model.
Tasks: Control of a simulated two-arm robot from camera images, as Anthropic describes it.
Setting: Max effort; Anthropic says this is the highest-available level and that it ran the evaluation internally.
Limits: Provider-reported by Anthropic in its launch announcement; not independently reproduced here. Anthropic points to the Fable 5.2 and Mythos 5.2 system card for methods; that card was not opened, so task subset, sample size, harness, prompts, trial counts and run dates are unverified. Results under the same benchmark name can be shown together as reported scores, but unmatched settings and missing methods rule out a controlled ranking. Run internally by Anthropic rather than by the benchmark owner.
Anthropic
