List your company

Group Booking with failures (Grand Rollback) on τ2-bench+ / REALM-bench+ P14 v1 (test)

2Result Success RateRAC
-0.080.4611.54May 5, 2026

Evaluation results

MethodLinks
RACFramework=RAC2026.052189740
SagaLLMFramework=SagaLLM2026.051951732
LGFramework=LG2026.050—180
LG(PE)Framework=LG(PE)2026.050—170