[RSS LABS]score: 0.41
EVA-Bench 2.0 Covers 121 Tools and 213 Agent Scenarios Across 3 Domains
June 4, 2026
ServiceNow released EVA-Bench Data 2.0, an agent evaluation benchmark spanning 3 domains, 121 tools, and 213 scenarios. No model scores or methodology details are available from the source.
HOW THIS AFFECTS YOU
●
builderWorth checking if you need a structured benchmark to evaluate tool-use agents before shipping.
●
researcherEVA-Bench 2.0 offers broader tool and scenario coverage for evaluating agent behavior across enterprise-relevant domains.