II.
TestSet overview
Reference · livetest-set:assistantbench-214-web-tasks
AssistantBench 214 web tasks overview
Canonical AssistantBench artifact for web-research and web-agent task completion evaluation.
Attributes
displayName
AssistantBench 214 web tasks
benchmarkId
caseCount
214
composition
AssistantBench's realistic web-agent tasks cover time-consuming
information-seeking and assistant workflows across many websites
and domains, with automatic evaluation.
homepageUrl
description
Canonical AssistantBench artifact for web-research and web-agent
task completion evaluation.
Outgoing edges
belongs_to_benchmark1
- benchmark:assistant-bench·BenchmarkAssistantBench
Incoming edges
None.