iiRecord
Agentic AI Atlas · eval-run:math.gpt-5.2025-08
eval-run:math.gpt-5.2025-08a5c.ai
II.
EvalRun JSON

eval-run:math.gpt-5.2025-08

Structured · live

eval-run:math.gpt-5.2025-08 json

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

File · benchmarks/eval-runs/extended-eval-runs.yamlCluster · benchmarks
Record JSON
{
  "id": "eval-run:math.gpt-5.2025-08",
  "_kind": "EvalRun",
  "_file": "benchmarks/eval-runs/extended-eval-runs.yaml",
  "_cluster": "benchmarks",
  "attributes": {
    "target": "model:gpt-5@current",
    "benchmarkId": "benchmark:math",
    "testSetId": "test-set:swe-bench-verified-2024-12",
    "targetId": "model:gpt-5@current",
    "runAt": "2025-08-07T00:00:00Z",
    "runBy": "openai",
    "configHash": "sha256:placeholder-gpt-5-math"
  },
  "outgoingEdges": [
    {
      "from": "eval-run:math.gpt-5.2025-08",
      "to": "model:gpt-5@current",
      "kind": "evaluates_target"
    },
    {
      "from": "eval-run:math.gpt-5.2025-08",
      "to": "benchmark:math",
      "kind": "for_benchmark"
    }
  ],
  "incomingEdges": [
    {
      "from": "eval-result:math.gpt-5.001",
      "to": "eval-run:math.gpt-5.2025-08",
      "kind": "belongs_to_eval_run"
    }
  ]
}