iiRecord
Agentic AI Atlas · AssistantBench 214 web tasks
test-set:assistantbench-214-web-tasksa5c.ai
II.
TestSet JSON

test-set:assistantbench-214-web-tasks

Structured · live

AssistantBench 214 web tasks json

Inspect the normalized record payload exactly as the atlas UI reads it.

File · benchmarks/test-sets/source-backed-benchmark-test-sets.yamlCluster · benchmarks
Record JSON
{
  "id": "test-set:assistantbench-214-web-tasks",
  "_kind": "TestSet",
  "_file": "benchmarks/test-sets/source-backed-benchmark-test-sets.yaml",
  "_cluster": "benchmarks",
  "attributes": {
    "displayName": "AssistantBench 214 web tasks",
    "benchmarkId": "benchmark:assistant-bench",
    "caseCount": 214,
    "composition": "AssistantBench's realistic web-agent tasks cover time-consuming\ninformation-seeking and assistant workflows across many websites\nand domains, with automatic evaluation.\n",
    "homepageUrl": "https://assistantbench.github.io/",
    "description": "Canonical AssistantBench artifact for web-research and web-agent\ntask completion evaluation.\n"
  },
  "outgoingEdges": [
    {
      "from": "test-set:assistantbench-214-web-tasks",
      "to": "benchmark:assistant-bench",
      "kind": "belongs_to_benchmark"
    }
  ],
  "incomingEdges": []
}