iiRecord
Agentic AI Atlas · MultiPL-E
benchmark:multipl-ea5c.ai
II.
Benchmark JSON

benchmark:multipl-e

Structured · live

MultiPL-E json

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

File · benchmarks/benchmarks/multipl-e.yamlCluster · benchmarks
Record JSON
{
  "id": "benchmark:multipl-e",
  "_kind": "Benchmark",
  "_file": "benchmarks/benchmarks/multipl-e.yaml",
  "_cluster": "benchmarks",
  "attributes": {
    "displayName": "MultiPL-E",
    "homepageUrl": "https://github.com/nuprl/MultiPL-E",
    "kind": "code-generation",
    "targetsKind": "ModelVersion",
    "description": "MultiPL-E — multi-language extension of HumanEval / MBPP across 18+ programming languages.\n"
  },
  "outgoingEdges": [],
  "incomingEdges": [
    {
      "from": "eval-result:multipl-e.codestral-25-01.001",
      "to": "benchmark:multipl-e",
      "kind": "scored_against",
      "attributes": {}
    },
    {
      "from": "eval-run:multipl-e.codestral-25-01.2025-01",
      "to": "benchmark:multipl-e",
      "kind": "for_benchmark"
    },
    {
      "from": "scope-boundary:multipl-e.scope",
      "to": "benchmark:multipl-e",
      "kind": "bounds_subject"
    }
  ]
}