iiRecord
Agentic AI Atlas · eval-run:gpqa-diamond.gemini-2-5-pro.2025-06
eval-run:gpqa-diamond.gemini-2-5-pro.2025-06a5c.ai
II.
EvalRun JSON

eval-run:gpqa-diamond.gemini-2-5-pro.2025-06

Structured · live

eval-run:gpqa-diamond.gemini-2-5-pro.2025-06 json

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

File · benchmarks/eval-runs/eval-runs-google.yamlCluster · benchmarks
Record JSON
{
  "id": "eval-run:gpqa-diamond.gemini-2-5-pro.2025-06",
  "_kind": "EvalRun",
  "_file": "benchmarks/eval-runs/eval-runs-google.yaml",
  "_cluster": "benchmarks",
  "attributes": {
    "target": "model:gemini-2-5-pro@current",
    "benchmarkId": "benchmark:gpqa",
    "testSetId": "test-set:gpqa-diamond-2024",
    "targetId": "model:gemini-2-5-pro@current",
    "runAt": "2025-06-17T00:00:00Z",
    "runBy": "google-deepmind",
    "configHash": "sha256:placeholder-gemini-2-5-pro-gpqa-diamond"
  },
  "outgoingEdges": [
    {
      "from": "eval-run:gpqa-diamond.gemini-2-5-pro.2025-06",
      "to": "model:gemini-2-5-pro@current",
      "kind": "evaluates_target"
    },
    {
      "from": "eval-run:gpqa-diamond.gemini-2-5-pro.2025-06",
      "to": "test-set:gpqa-diamond-2024",
      "kind": "uses_test_set"
    },
    {
      "from": "eval-run:gpqa-diamond.gemini-2-5-pro.2025-06",
      "to": "benchmark:gpqa",
      "kind": "for_benchmark"
    }
  ],
  "incomingEdges": [
    {
      "from": "eval-result:gpqa-diamond.gemini-2-5-pro.001",
      "to": "eval-run:gpqa-diamond.gemini-2-5-pro.2025-06",
      "kind": "belongs_to_eval_run"
    }
  ]
}