Fetching from the wire…
01
02
03
04
05
06
07
08
Research2026-08-14 · source-backed
VAKRA (arXiv 2608.12282) benchmarks agents against 8,000+ executable APIs across 62 domains, verifying by re-executing predicted calls against live endpoints. Accuracy falls to 50-51% on compositional APIs and degrades over 50% as depth grows. Failures concentrate in entity disambiguation and cross-source grounding, not tool invocation mechanics, so retries and better function schemas won't move these numbers.
Each link below shares sources, entities, or timing with this story.
Shared entity: Best / Same source domain / Shared topic / Earlier coverage / Tension
Both cover Best; reported by the same outlet (arxiv.org); overlapping topics (against, agent, best, better).
Shared entity: Accuracy / Same source domain / Shared topic / Earlier coverage
Both cover Accuracy; reported by the same outlet (arxiv.org); overlapping topics (accuracy, benchmark, best, call, depth).
Shared entity: Best / Same source domain / Shared topic / Earlier coverage
Both cover Best; reported by the same outlet (arxiv.org); overlapping topics (against, agent, benchmark, best, call).
Shared entity: Accuracy / Same source domain / Shared topic / Earlier coverage / Tension
Both cover Accuracy; reported by the same outlet (arxiv.org); overlapping topics (accuracy, agent).
Shared entity: APIs / Same source domain / Shared topic / Earlier coverage / Tension
Both cover APIs; reported by the same outlet (arxiv.org); overlapping topics (against, apis).
Shared entity: Accuracy / Shared topic / Earlier coverage / Tension
Both cover Accuracy; overlapping topics (accuracy, against, agent, benchmark); earlier Accuracy coverage from 2026-05-10.
Shared entity: Best / Same source domain / Shared topic / Earlier coverage / Downstream implication
Both cover Best; reported by the same outlet (arxiv.org); overlapping topics (agent, best).
Shared entity: Best / Same source domain / Shared topic / Earlier coverage
Both cover Best; reported by the same outlet (arxiv.org); overlapping topics (agent, best, call).