{"record": {"aliases": [], "artifacts": [], "categories": ["safety", "healthcare", "vision"], "description": {"en": "FigQA is a multiple-choice benchmark on interpreting scientific figures from biology papers. It evaluates dual-use biological knowledge and multimodal reasoning relevant to bioweapons development."}, "documents": [{"document_type": "registry_page", "id": "llm_stats:https://api.zeroeval.com/leaderboard/benchmarks/figqa?top_n=500", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/figqa?top_n=500", "title": "FigQA"}], "evidence_summary": {"document_count": 1, "model_count": 3, "model_count_basis": "source_model_id"}, "key": "llm-stats:figqa", "modality": "multimodal", "name": "FigQA", "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/figqa?top_n=500"}, "publisher": null, "released": null, "released_reference": null, "schema_version": 1, "sizes": [], "slug": "llm-stats-figqa", "source": "llm_stats", "source_benchmark_id": "figqa"}, "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/figqa?top_n=500", "key": "llm-stats:figqa", "measured_by": null, "model_id": "claude-mythos-preview", "model_name": "Claude Mythos Preview", "obs_id": "llm_stats:figqa:claude-mythos-preview", "organization": "Anthropic", "rank_in_source_response": 1, "raw_value": "0.89", "reported_by": "self_reported", "reported_date": "2026-04-07", "series_id": "llm_stats:figqa:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/figqa?top_n=500", "value": 0.89, "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/figqa?top_n=500", "key": "llm-stats:figqa", "measured_by": null, "model_id": "claude-opus-4-6", "model_name": "Claude Opus 4.6", "obs_id": "llm_stats:figqa:claude-opus-4-6", "organization": "Anthropic", "rank_in_source_response": 2, "raw_value": "0.783", "reported_by": "self_reported", "reported_date": "2026-02-05", "series_id": "llm_stats:figqa:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/figqa?top_n=500", "value": 0.783, "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/figqa?top_n=500", "key": "llm-stats:figqa", "measured_by": null, "model_id": "grok-4.1-thinking-2025-11-17", "model_name": "Grok-4.1 Thinking", "obs_id": "llm_stats:figqa:grok-4.1-thinking-2025-11-17", "organization": "xAI", "rank_in_source_response": 3, "raw_value": "0.34", "reported_by": "self_reported", "reported_date": "2025-11-17", "series_id": "llm_stats:figqa:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/figqa?top_n=500", "value": 0.34, "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:figqa:grok-4.1-thinking-2025-11-17", "reported_at": "2025-11-17", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/figqa?top_n=500"}, "first_score_report": null, "key": "llm-stats:figqa", "max_score_contradicted": false, "metric": null, "observation_count": 3, "observed_max": 0.89, "score_summary": {"display_max": 89.0, "display_multiplier": 100, "model_count": 3, "model_count_basis": "source_model_id", "numeric_count": 3, "raw_max": 0.89, "raw_min": 0.34, "source_reference": {"crawled_at": "2026-08-17T18:49:35.489925+00:00", "obs_id": "llm_stats:figqa:claude-mythos-preview", "reported_date": "2026-04-07", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/figqa?top_n=500"}, "unit": null}, "series_id": "llm_stats:figqa:default"}}}, "siblings": []}
