{"record": {"aliases": [], "artifacts": [], "categories": ["long_context", "reasoning"], "description": {"en": "Flexible Length Question Answering dataset for evaluating the impact of input length on reasoning performance of language models, featuring True/False questions embedded in contexts of varying lengths (250-3000 tokens) across three reasoning tasks: Monotone Relations, People In Rooms, and simplified Ruletaker"}, "documents": [{"document_type": "registry_page", "id": "llm_stats:https://api.zeroeval.com/leaderboard/benchmarks/flenqa?top_n=500", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/flenqa?top_n=500", "title": "FlenQA"}], "evidence_summary": {"document_count": 1, "model_count": 2, "model_count_basis": "source_model_id"}, "key": "llm-stats:flenqa", "modality": "text", "name": "FlenQA", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"crawl_bundle": "llm_stats_2026-08-17", "crawled_at": "2026-08-17T18:49:35.489925+00:00", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/flenqa?top_n=500"}, "publisher": null, "released": null, "released_reference": null, "schema_version": 1, "sizes": [], "slug": "llm-stats-flenqa", "source": "llm_stats", "source_benchmark_id": "flenqa"}, "schema_version": 1, "scores_by_source": {"llm_stats": {"rows": [{"comparable_group": null, "crawled_at": "2026-08-17T18:49:35.489925+00:00", "date_precision": "model_announcement", "document_id": "llm_stats:https://api.zeroeval.com/leaderboard/benchmarks/flenqa?top_n=500", "key": "llm-stats:flenqa", "measured_by": null, "model_id": "phi-4-reasoning-plus", "model_name": "Phi 4 Reasoning Plus", "obs_id": "llm_stats:flenqa:phi-4-reasoning-plus", "organization": "Microsoft", "rank_in_source_response": 1, "raw_value": "0.979", "reported_by": "self_reported", "reported_date": "2025-04-30", "series_id": "llm_stats:flenqa:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/flenqa?top_n=500", "value": 0.979, "value_kind": "number"}, {"comparable_group": null, "crawled_at": "2026-08-17T18:49:35.489925+00:00", "date_precision": "model_announcement", "document_id": "llm_stats:https://api.zeroeval.com/leaderboard/benchmarks/flenqa?top_n=500", "key": "llm-stats:flenqa", "measured_by": null, "model_id": "phi-4-reasoning", "model_name": "Phi 4 Reasoning", "obs_id": "llm_stats:flenqa:phi-4-reasoning", "organization": "Microsoft", "rank_in_source_response": 2, "raw_value": "0.977", "reported_by": "self_reported", "reported_date": "2025-04-30", "series_id": "llm_stats:flenqa:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/flenqa?top_n=500", "value": 0.977, "value_kind": "number"}], "series": {"bounds": {"basis": "aggregator_declared", "max": 1.0, "min": null}, "declared_max": 1.0, "direction": "higher_is_better", "direction_basis": "source_rank_descending", "display_scale": null, "first_score_record": {"date_precision": "model_announcement", "obs_id": "llm_stats:flenqa:phi-4-reasoning", "reported_at": "2025-04-30", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/flenqa?top_n=500"}, "first_score_report": null, "key": "llm-stats:flenqa", "max_score_contradicted": false, "metric": null, "observation_count": 2, "observed_max": 0.979, "score_summary": {"display_max": 97.89999999999999, "display_multiplier": 100, "model_count": 2, "model_count_basis": "source_model_id", "numeric_count": 2, "raw_max": 0.979, "raw_min": 0.977, "source_reference": {"crawled_at": "2026-08-17T18:49:35.489925+00:00", "obs_id": "llm_stats:flenqa:phi-4-reasoning-plus", "reported_date": "2025-04-30", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/flenqa?top_n=500"}, "unit": null}, "series_id": "llm_stats:flenqa:default"}}}, "siblings": []}
