List your company

Long-horizon task execution on Red Dead Redemption 2 (RDR2)

56.4Success Rate (SR)SPIKE
5.75218.90132.0545.199May 18, 2026

Evaluation results

MethodLinks
SPIKE2026.0556.43,40054.1
CRADLE2026.0543.66,600102.2
StarDojo2026.0533.35,70067.5
Reflexion-like2026.0517.93,50041.4
ReAct-like2026.0515.42,10028.2
Voyager-like2026.057.74,80072.4