A test run is a single execution of a benchmark test using a defined model configuration.
Each run represents how a particular large language model (LLM) — such as GPT-4, Claude-3, or Gemini — performed on a given task at a specific time, with specific settings.
A test run includes:
Together, test runs make it possible to compare models, providers, and configurations across benchmarks in a transparent and reproducible way.
book-page information-extraction printed 20 en list bibliographic
openai gpt-4.1-mini · temp 0.0 · dataclass Document
One model response from this run. The per-input comparison needs JavaScript.
book-page information-extraction printed 20 en list bibliographic
openai gpt-5-mini · temp 0.0 · dataclass Document
One model response from this run. The per-input comparison needs JavaScript.
book-page information-extraction printed 20 en list bibliographic
genai gemini-2.5-flash-lite · temp 0.0 · dataclass Document
One model response from this run. The per-input comparison needs JavaScript.
{
"metadata": {
"title": "BIBLIOGRAPHY",
"year": "1947",
"page_number": 9
},
"entries": [
{
"id": "140",
"type": "other",
"title": "Smellie, K. B., Why We Read History",
"container_title": null,
"author": [
{
"family": "Smellie",
"given": "K. B."
}
],
"note": "London: P. Elek.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "141",
"type": "other",
"title": "Toynbee, Arnold J., A Study of History",
"container_title": null,
"author": [
{
"family": "Toynbee",
"given": "Arnold J."
}
],
"note": "abridged by D. C. Somervell, New York: Oxford (Abridgement of vols. I-VI of A Study of History).",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "142",
"type": "other",
"title": "Verberne, L. G. J., In den spiegel van het verleden: historische opstellen",
"container_title": null,
"author": [
{
"family": "Verberne",
"given": "L. G. J."
}
],
"note": "Utrecht, Antwerpen: Bruna.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "143",
"type": "other",
"title": "Vevir, José, Esquemas de metodologia eclesiástica",
"container_title": null,
"author": [
{
"family": "Vevir",
"given": "José"
}
],
"note": "Barcelona: Instituto.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "144",
"type": "other",
"title": "Walz, Hans H., Sinn und Ziel der Geschichte",
"container_title": null,
"author": [
{
"family": "Walz",
"given": "Hans H."
}
],
"note": "Bonn: Universitäts-Verlag.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "145",
"type": "other",
"title": "Atti del congresso internazionale di filosofia 15-20 November, 1946",
"container_title": null,
"author": [],
"note": "Milano: Castellani.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "146",
"type": "journal-article",
"title": "\"Storia e materialismo storico\"",
"container_title": null,
"author": [
{
"family": "Drago",
"given": "Pietro C."
}
],
"note": "vol. I, 109-117.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "147",
"type": "journal-article",
"title": "\"Critica del giudizio storico\"",
"container_title": null,
"author": [
{
"family": "Fazio-Allmayer",
"given": "Vito"
}
],
"note": "vol. I, 281-290.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "148",
"type": "journal-article",
"title": "\"Forza e spirito nella storia\"",
"container_title": null,
"author": [
{
"family": "Nyman",
"given": "A."
}
],
"note": "vol. I, 191-200.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "149",
"type": "journal-article",
"title": "\"Sur le concept d'histoire\"",
"container_title": null,
"author": [
{
"family": "Benjamin",
"given": "Walter"
}
],
"note": "Les temps modernes II, 623-634.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "150",
"type": "book",
"title": "\"Philosophy of History\"",
"container_title": null,
"author": [
{
"family": "Boodin",
"given": "John E."
}
],
"note": "in Dagobert D. Runes, ed., 20th Century Philosophy, New York: Philosophical Library, 89-106.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "151",
"type": "book",
"title": "\"De la perception à l'histoire. La philosophie de M. Merleau-Ponty\"",
"container_title": null,
"author": [
{
"family": "Caillois",
"given": "R. P."
}
],
"note": "Deucalion II, 57-85.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "152",
"type": "book",
"title": "\"Commemorazione di Adolfo Omodeo\"",
"container_title": null,
"author": [
{
"family": "Cantimori",
"given": "M."
},
{
"family": "Delio",
"given": ""
}
],
"note": "Annali della Scuola Normale di Pisa, fasc. 3-4; (reprinted in D. Cantimori, Studi di storia, Torino: Einaudi, 1959).",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "153",
"type": "book",
"title": "\"The Historian and the Philosophy of Science\"",
"container_title": null,
"author": [
{
"family": "Cowan",
"given": "Thomas"
}
],
"note": "Isis XXXVIII, 11-18.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "154",
"type": "book",
"title": "\"History as a Science\"",
"container_title": null,
"author": [
{
"family": "Crawford",
"given": "R. M."
}
],
"note": "Historical Studies: Australia and New Zealand III, 153-175.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "155",
"type": "journal-article",
"title": "\"Christianisme et histoire\"",
"container_title": null,
"author": [
{
"family": "Daniélou",
"given": "J."
}
],
"note": "Etudes CCLIV, 166-184.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "156",
"type": "journal-article",
"title": "\"Wahrheit der Geschichte. Geschichte der Wahrheit\"",
"container_title": null,
"author": [
{
"family": "Dell",
"given": "Stanley"
}
],
"note": "Amerikanische Rundschau III, 44-53.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "157",
"type": "journal-article",
"title": "\"The Idea of Progress\"",
"container_title": null,
"author": [
{
"family": "Fay",
"given": "Sidney B."
}
],
"note": "American Historical Review LII, 231-246.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "158",
"type": "journal-article",
"title": "\"Théologie et histoire\"",
"container_title": null,
"author": [
{
"family": "Fessard",
"given": "Gaston"
}
],
"note": "Dieu vivant VIII, 39-65.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "159",
"type": "journal-article",
"title": "\"Acton, Creighton, and Lea: a Study in History and Ethics\"",
"container_title": null,
"author": [
{
"family": "Fish",
"given": "Andrew"
}
],
"note": "Pacific Historical Review XVI, 59-69.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": false
},
{
"id": "160",
"type": "journal-article",
"title": "\"Le conscience historique en Italie\"",
"container_title": null,
"author": [
{
"family": "Garosci",
"given": "A."
}
],
"note": "Les temps modernes III, 23-24.",
"publisher": null,
"publisher_place": null,
"issued": null,
"page": null,
"incomplete": true
}
]
}
book-page information-extraction printed 20 en list bibliographic
openai gpt-4.1 · temp 0.0 · dataclass Document
One model response from this run. The per-input comparison needs JavaScript.
book-page information-extraction printed 20 en list bibliographic
openai gpt-5 · temp 0.0 · dataclass Document
One model response from this run. The per-input comparison needs JavaScript.
book-page information-extraction printed 20 en list bibliographic
genai gemini-2.5-pro · temp 0.0 · dataclass Document
One model response from this run. The per-input comparison needs JavaScript.
{
"metadata": {
"title": "BIBLIOGRAPHY",
"year": "1947",
"page_number": 9
},
"entries": [
{
"id": "140",
"type": "book",
"title": "Why We Read History",
"author": [
{
"family": "Smellie",
"given": "K. B."
}
],
"publisher": "P. Elek",
"publisher_place": "London"
},
{
"id": "141",
"type": "book",
"title": "A Study of History",
"author": [
{
"family": "Toynbee",
"given": "Arnold J."
}
],
"note": "abridged by D. C. Somervell. New York: Oxford (Abridegement of vols. I-VI of A Study of History).",
"publisher": "Oxford",
"publisher_place": "New York"
},
{
"id": "142",
"type": "book",
"title": "In den spiegel van het verleden: historische opstellen",
"author": [
{
"family": "Verberne",
"given": "L. G. J."
}
],
"publisher": "Bruna",
"publisher_place": "Utrecht, Antwerpen"
},
{
"id": "143",
"type": "book",
"title": "Esquemas de metodologia eclesiástica",
"author": [
{
"family": "Vevir",
"given": "José"
}
],
"publisher": "Instituto",
"publisher_place": "Barcelona"
},
{
"id": "144",
"type": "book",
"title": "Sinn und Ziel der Geschichte",
"author": [
{
"family": "Walz",
"given": "Hans H."
}
],
"publisher": "Universitäts-Verlag",
"publisher_place": "Bonn"
},
{
"id": "145",
"type": "book",
"title": "Atti del congresso internazionale di filosofia 15-20 November, 1946",
"note": "Relevant articles:",
"publisher": "Castellani",
"publisher_place": "Milano"
},
{
"id": "146",
"type": "journal-article",
"title": "Storia e materialismo storico",
"author": [
{
"family": "Drago",
"given": "Pietro C."
}
],
"relation": "145",
"volume": "I",
"page": "109-117"
},
{
"id": "147",
"type": "journal-article",
"title": "Critica del giudizio storico",
"author": [
{
"family": "Fazio-Allmayer",
"given": "Vito"
}
],
"relation": "145",
"volume": "I",
"page": "281-290"
},
{
"id": "148",
"type": "journal-article",
"title": "Forza e spirito nella storia",
"author": [
{
"family": "Nyman",
"given": "A."
}
],
"relation": "145",
"volume": "I",
"page": "191-200"
},
{
"id": "149",
"type": "journal-article",
"title": "Sur le concept d'histoire",
"container_title": "Les temps modernes II",
"author": [
{
"family": "Benjamin",
"given": "Walter"
}
],
"page": "623-634"
},
{
"id": "150",
"type": "journal-article",
"title": "Philosophy of History",
"container_title": "20th Century Philosophy",
"author": [
{
"family": "Boodin",
"given": "John E."
}
],
"publisher": "Philosophical Library",
"editor": [
"Dagobert D. Runes"
],
"publisher_place": "New York",
"page": "89-106"
},
{
"id": "151",
"type": "journal-article",
"title": "De la perception à l'histoire. La philosophie de M. Merleau-Ponty",
"container_title": "Deucalion II",
"author": [
{
"family": "Caillois",
"given": "R. P."
}
],
"page": "57-85"
},
{
"id": "152",
"type": "journal-article",
"title": "Commemorazione di Adolfo Omodeo",
"container_title": "Annali della Scuola Normale di Pisa",
"author": [
{
"family": "Cantimori",
"given": "Delio"
}
],
"note": "(reprinted in D. Cantimori, Studi di storia, Torino: Einaudi, 1959).",
"fascicle": "3-4"
},
{
"id": "153",
"type": "journal-article",
"title": "The Historian and the Philosophy of Science",
"container_title": "Isis XXXVIII",
"author": [
{
"family": "Cowan",
"given": "Thomas"
}
],
"page": "11-18"
},
{
"id": "154",
"type": "journal-article",
"title": "History as a Science",
"container_title": "Historical Studies: Australia and New Zealand III",
"author": [
{
"family": "Crawford",
"given": "R. M."
}
],
"page": "153-175"
},
{
"id": "155",
"type": "journal-article",
"title": "Christianisme et histoire",
"container_title": "Etudes CCLIV",
"author": [
{
"family": "Daniélou",
"given": "J."
}
],
"page": "166-184"
},
{
"id": "156",
"type": "journal-article",
"title": "Wahrheit der Geschichte. Geschichte der Wahrheit",
"container_title": "Amerikanische Rundschau III",
"author": [
{
"family": "Dell",
"given": "Stanley"
}
],
"page": "44-53"
},
{
"id": "157",
"type": "journal-article",
"title": "The Idea of Progress",
"container_title": "American Historical Review LII",
"author": [
{
"family": "Fay",
"given": "Sidney B."
}
],
"page": "231-246"
},
{
"id": "158",
"type": "journal-article",
"title": "Théologie et histoire",
"container_title": "Dieu vivant VIII",
"author": [
{
"family": "Fessard",
"given": "Gaston"
}
],
"page": "39-65"
},
{
"id": "159",
"type": "journal-article",
"title": "Acton, Creighton, and Lea: a Study in History and Ethics",
"container_title": "Pacific Historical Review XVI",
"author": [
{
"family": "Fish",
"given": "Andrew"
}
],
"page": "59-69"
},
{
"id": "160",
"type": "journal-article",
"title": "Le conscience historique en Italie",
"container_title": "Les temps modernes III",
"author": [
{
"family": "Garosci",
"given": "A."
}
],
"page": "23-24"
}
]
}
book-page information-extraction printed 20 en list bibliographic
openai gpt-5-nano · temp 0.0 · dataclass Document
One model response from this run. The per-input comparison needs JavaScript.
book-page information-extraction printed 20 en list bibliographic
openai gpt-4o · temp 0.0 · dataclass Document
One model response from this run. The per-input comparison needs JavaScript.
book-page information-extraction printed 20 en list bibliographic
genai gemini-2.5-flash · temp 0.0 · dataclass Document
One model response from this run. The per-input comparison needs JavaScript.
{"metadata": {"title": "Bibliography and Articles", "year": "1947", "page_number": 9}, "entries": [{"id": "140", "type": "book", "title": "Why We Read History", "author": [{"family": "Smellie", "given": "K. B."}], "publisher": "P. Elek", "publisher_place": "London"}, {"id": "141", "type": "book", "title": "A Study of History", "author": [{"family": "Toynbee", "given": "Arnold J."}], "note": "abridged by D. C. Somervell (Abridgement of vols. I-VI of A Study of History)", "publisher": "Oxford", "publisher_place": "New York"}, {"id": "142", "type": "book", "title": "In den spiegel van het verleden: historische opstellen", "author": [{"family": "Verberne", "given": "L. G. J."}], "publisher": "Bruna", "publisher_place": "Utrecht, Antwerpen"}, {"id": "143", "type": "book", "title": "Esquemas de metodologia eclesiástica", "author": [{"family": "Veivir", "given": "José"}], "publisher": "Instituto", "publisher_place": "Barcelona"}, {"id": "144", "type": "book", "title": "Sinn und Ziel der Geschichte", "author": [{"family": "Walz", "given": "Hans H."}], "publisher": "Universitäts-Verlag", "publisher_place": "Bonn"}, {"id": "145", "type": "other", "title": "Atti del congresso internazionale di filosofia", "event_date": "15-20 November, 1946", "related": ["146", "147", "148", "149"], "page": "Relevant articles:", "reprint": "Milano: Castellani"}, {"id": "146", "type": "journal-article", "title": "Storia e materialismo storico", "author": [{"family": "Drago", "given": "Pietro C."}], "volume": "I", "page": "109-117"}, {"id": "147", "type": "journal-article", "title": "Critica del giudizio storico", "author": [{"family": "Fazio-Allmayer", "given": "Vito"}], "volume": "I", "page": "281-290"}, {"id": "148", "type": "journal-article", "title": "Forza e spirito nella storia", "author": [{"family": "Nyman", "given": "A."}], "volume": "I", "page": "191-200"}, {"id": "149", "type": "journal-article", "title": "Sur le concept d'histoire", "author": [{"family": "Benjamin", "given": "Walter"}], "volume": "II", "page": "623-634", "reprint": "Les temps modernes"}, {"id": "150", "type": "journal-article", "title": "Philosophy of History", "container_title": "20th Century Philosophy", "author": [{"family": "Boodin", "given": "John E."}], "editor": ["Dagobert D. Runes"], "publisher_place": "New York", "page": "89-106", "reprint": "Philosophical Library"}, {"id": "151", "type": "journal-article", "title": "De la perception à l'histoire. La philosophie de M. Merleau-Ponty", "container_title": "Deucalion", "author": [{"family": "Caillois", "given": "R. P."}], "volume": "II", "page": "57-85"}, {"id": "152", "type": "journal-article", "title": "Commemorazione di Adolfo Omodeo", "container_title": "Annali della Scuola Normale di Pisa", "author": [{"family": "Cantimori", "given": "Delio"}], "note": "reprinted in D. Cantimori, Studi di storia, Torino: Einaudi, 1959", "fascicle": "3-4"}, {"id": "153", "type": "journal-article", "title": "The Historian and the Philosophy of Science", "container_title": "Isis", "author": [{"family": "Cowan", "given": "Thomas"}], "volume": "XXXVIII", "page": "11-18"}, {"id": "154", "type": "journal-article", "title": "History as a Science", "container_title": "Historical Studies: Australia and New Zealand", "author": [{"family": "Crawford", "given": "R. M."}], "volume": "III", "page": "153-175"}, {"id": "155", "type": "journal-article", "title": "Christianisme et histoire", "container_title": "Etudes", "author": [{"family": "Daniélou", "given": "J."}], "volume": "CCLIV", "page": "166-184"}, {"id": "156", "type": "journal-article", "title": "Wahrheit der Geschichte. Geschichte der Wahrheit", "container_title": "Amerikanische Rundschau", "author": [{"family": "Dell", "given": "Stanley"}], "volume": "III", "page": "44-53"}, {"id": "157", "type": "journal-article", "title": "The Idea of Progress", "container_title": "American Historical Review", "author": [{"family": "Fay", "given": "Sidney B."}], "volume": "LII", "page": "231-246"}, {"id": "158", "type": "journal-article", "title": "Théologie et histoire", "container_title": "Dieu vivant", "author": [{"family": "Fessard", "given": "Gaston"}], "volume": "VIII", "page": "39-65"}, {"id": "159", "type": "journal-article", "title": "Acton, Creighton and Lea: a Study in History and Ethics", "container_title": "Pacific Historical Review", "author": [{"family": "Fish", "given": "Andrew"}], "volume": "XVI", "page": "59-69"}, {"id": "160", "type": "journal-article", "title": "Le conscience historique en Italie", "container_title": "Les temps modernes", "author": [{"family": "Garosci", "given": "A."}], "volume": "III", "page": "23-24", "incomplete": null}]}
book-page information-extraction printed 20 en list bibliographic
openai gpt-4o-mini · temp 0.0 · dataclass Document
One model response from this run. The per-input comparison needs JavaScript.