{"record": {"aliases": [], "artifacts": [], "categories": ["multimodal", "reasoning", "structured_output", "vision"], "description": {"en": "CharXiv-D is the descriptive questions subset of the CharXiv benchmark, designed to assess multimodal large language models' ability to extract basic information from scientific charts. It contains descriptive questions covering information extraction, enumeration, pattern recognition, and counting across 2,323 diverse charts from arXiv papers, all curated and verified by human experts."}, "documents": [{"document_type": "registry_page", "id": "llm_stats:https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "title": "CharXiv-D"}], "evidence_summary": {"document_count": 1, "model_count": 17, "model_count_basis": "source_model_id"}, "key": "llm-stats:charxiv-d", "modality": "multimodal", "name": "CharXiv-D", "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/charxiv-d?top_n=500"}, "publisher": null, "released": null, "released_reference": null, "schema_version": 1, "sizes": [], "slug": "llm-stats-charxiv-d", "source": "llm_stats", "source_benchmark_id": "charxiv-d"}, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "seed-2.1-pro", "model_name": "Seed 2.1 Pro", "obs_id": "llm_stats:charxiv-d:seed-2.1-pro", "organization": "ByteDance", "rank_in_source_response": 1, "raw_value": "0.955", "reported_by": "self_reported", "reported_date": "2026-06-24", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.955, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "seed-2.1-turbo", "model_name": "Seed 2.1 Turbo", "obs_id": "llm_stats:charxiv-d:seed-2.1-turbo", "organization": "ByteDance", "rank_in_source_response": 2, "raw_value": "0.946", "reported_by": "self_reported", "reported_date": "2026-06-24", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.946, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "qwen3-vl-32b-instruct", "model_name": "Qwen3 VL 32B Instruct", "obs_id": "llm_stats:charxiv-d:qwen3-vl-32b-instruct", "organization": "Qwen", "rank_in_source_response": 3, "raw_value": "0.905", "reported_by": "self_reported", "reported_date": "2025-09-22", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.905, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "qwen3-vl-32b-thinking", "model_name": "Qwen3 VL 32B Thinking", "obs_id": "llm_stats:charxiv-d:qwen3-vl-32b-thinking", "organization": "Qwen", "rank_in_source_response": 4, "raw_value": "0.902", "reported_by": "self_reported", "reported_date": "2025-09-22", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.902, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "gpt-4.5", "model_name": "GPT-4.5", "obs_id": "llm_stats:charxiv-d:gpt-4.5", "organization": "OpenAI", "rank_in_source_response": 5, "raw_value": "0.9", "reported_by": "self_reported", "reported_date": "2025-02-27", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.9, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "gpt-4.1-mini-2025-04-14", "model_name": "GPT-4.1 mini", "obs_id": "llm_stats:charxiv-d:gpt-4.1-mini-2025-04-14", "organization": "OpenAI", "rank_in_source_response": 6, "raw_value": "0.884", "reported_by": "self_reported", "reported_date": "2025-04-14", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.884, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "command-a-plus-05-2026", "model_name": "Command A+", "obs_id": "llm_stats:charxiv-d:command-a-plus-05-2026", "organization": "Cohere", "rank_in_source_response": 7, "raw_value": "0.88", "reported_by": "self_reported", "reported_date": "2026-05-20", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.88, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "gpt-4.1-2025-04-14", "model_name": "GPT-4.1", "obs_id": "llm_stats:charxiv-d:gpt-4.1-2025-04-14", "organization": "OpenAI", "rank_in_source_response": 8, "raw_value": "0.879", "reported_by": "self_reported", "reported_date": "2025-04-14", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.879, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "qwen3-vl-30b-a3b-thinking", "model_name": "Qwen3 VL 30B A3B Thinking", "obs_id": "llm_stats:charxiv-d:qwen3-vl-30b-a3b-thinking", "organization": "Qwen", "rank_in_source_response": 9, "raw_value": "0.869", "reported_by": "self_reported", "reported_date": "2025-09-22", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.869, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "qwen3-vl-8b-thinking", "model_name": "Qwen3 VL 8B Thinking", "obs_id": "llm_stats:charxiv-d:qwen3-vl-8b-thinking", "organization": "Qwen", "rank_in_source_response": 10, "raw_value": "0.859", "reported_by": "self_reported", "reported_date": "2025-09-22", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.859, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "qwen3-vl-30b-a3b-instruct", "model_name": "Qwen3 VL 30B A3B Instruct", "obs_id": "llm_stats:charxiv-d:qwen3-vl-30b-a3b-instruct", "organization": "Qwen", "rank_in_source_response": 11, "raw_value": "0.855", "reported_by": "self_reported", "reported_date": "2025-09-22", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.855, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "gpt-4o-2024-08-06", "model_name": "GPT-4o", "obs_id": "llm_stats:charxiv-d:gpt-4o-2024-08-06", "organization": "OpenAI", "rank_in_source_response": 12, "raw_value": "0.853", "reported_by": "self_reported", "reported_date": "2024-08-06", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.853, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "qwen3-vl-4b-thinking", "model_name": "Qwen3 VL 4B Thinking", "obs_id": "llm_stats:charxiv-d:qwen3-vl-4b-thinking", "organization": "Qwen", "rank_in_source_response": 13, "raw_value": "0.839", "reported_by": "self_reported", "reported_date": "2025-09-22", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.839, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "qwen3-vl-8b-instruct", "model_name": "Qwen3 VL 8B Instruct", "obs_id": "llm_stats:charxiv-d:qwen3-vl-8b-instruct", "organization": "Qwen", "rank_in_source_response": 14, "raw_value": "0.83", "reported_by": "self_reported", "reported_date": "2025-09-22", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.83, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "qwen3-vl-4b-instruct", "model_name": "Qwen3 VL 4B Instruct", "obs_id": "llm_stats:charxiv-d:qwen3-vl-4b-instruct", "organization": "Qwen", "rank_in_source_response": 15, "raw_value": "0.762", "reported_by": "self_reported", "reported_date": "2025-09-22", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.762, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "gpt-4.1-nano-2025-04-14", "model_name": "GPT-4.1 nano", "obs_id": "llm_stats:charxiv-d:gpt-4.1-nano-2025-04-14", "organization": "OpenAI", "rank_in_source_response": 16, "raw_value": "0.739", "reported_by": "self_reported", "reported_date": "2025-04-14", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.739, "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/charxiv-d?top_n=500", "key": "llm-stats:charxiv-d", "measured_by": null, "model_id": "north-micro-vision-instruct", "model_name": "North Micro Vision Instruct", "obs_id": "llm_stats:charxiv-d:north-micro-vision-instruct", "organization": "Cohere", "rank_in_source_response": 17, "raw_value": "0.6", "reported_by": "self_reported", "reported_date": "2026-08-12", "series_id": "llm_stats:charxiv-d:default", "source": "llm_stats", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500", "value": 0.6, "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:charxiv-d:gpt-4o-2024-08-06", "reported_at": "2024-08-06", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500"}, "first_score_report": null, "key": "llm-stats:charxiv-d", "max_score_contradicted": false, "metric": null, "observation_count": 17, "observed_max": 0.955, "score_summary": {"display_max": 95.5, "display_multiplier": 100, "model_count": 17, "model_count_basis": "source_model_id", "numeric_count": 17, "raw_max": 0.955, "raw_min": 0.6, "source_reference": {"crawled_at": "2026-08-17T18:49:35.489925+00:00", "obs_id": "llm_stats:charxiv-d:seed-2.1-pro", "reported_date": "2026-06-24", "source_url": "https://api.zeroeval.com/leaderboard/benchmarks/charxiv-d?top_n=500"}, "unit": null}, "series_id": "llm_stats:charxiv-d:default"}}}, "siblings": []}
