Benchmark plan / mmmlu

MMMLU — OpenAI

knowledge · instruct · selected languages; aggregation pending

planned · No sub-1B cohort score yet. Planning cost W; actual GPU time is not measured.

Select languages and aggregation explicitly. Liquid LFM2 reports seven languages; do not silently pool all MMMLU languages.

Primary benchmark source ↗

Before scheduling

Pin dataset revision, split and item IDs; validate the native-model adapter; set context and generation budgets; freeze prompts, decoding, scoring and any grader/environment. Report every compatible system in the same scope. Long-context or gated data may require further preparation.

Presets: liquid-reference. Qualified detail-page visits affect planning priority, not automatic budget expansion.

Ontology record

{
  "number": 10,
  "benchmark_id": "mmmlu",
  "name": "MMMLU — OpenAI",
  "benchmark_family": "mmmlu",
  "task_ids": [
    "knowledge"
  ],
  "group": "liquid-reference",
  "source": "https://huggingface.co/datasets/openai/MMMLU",
  "liquid": [
    "LFM2"
  ],
  "presets": [
    "liquid-reference"
  ],
  "languages": [
    "selected languages; aggregation pending"
  ],
  "base_or_instruct": "instruct",
  "metric": "accuracy_by_language",
  "direction": "max",
  "dataset_revision": null,
  "split": null,
  "item_ids": null,
  "protocol_id": null,
  "generation_and_context_budget": {
    "max_input_tokens": null,
    "max_output_tokens": null,
    "sample_count": null
  },
  "grader_or_environment": "Deterministic scoring adapter to validate",
  "measured_cost": null,
  "planning_cost": "W",
  "status": "planned",
  "note": "Select languages and aggregation explicitly. Liquid LFM2 reports seven languages; do not silently pool all MMMLU languages.",
  "existing_adapter_task": null,
  "comparison_scope": "No sub-1B cohort result for this catalogue entry yet",
  "execution_tier": 2,
  "execution_priority_reason": "Liquid reference coverage",
  "execution_order": 110,
  "visits_30d": 0,
  "demand": 0,
  "priority": 0
}