{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/benchmarks/superglue/","edition":"2026-W40","date":"2026-09-28","echelle":"score_publie est le score de la source, dans son unité. score_0_1 est ce score ramené entre 0 et 1 pour le calcul. niveau_suggere est le score IA que cette seule mesure indique : 100 pour le meilleur modèle au lancement du site, 50 pour GPT-4o de novembre 2024. difficulte est le score IA d'un modèle qui obtient la moitié (score_0_1 = 0,5), pente la raideur de la courbe.","licence":"Réutilisation libre en citant « Quelle IA, quelleia.com » et la date de l'édition. Les mesures renvoient à leurs sources d'origine.","benchmark":{"id":"superglue","nom":"SuperGLUE","tache":"autre","etat":"archive","echelle":"unite","modeles":8,"poids_general":0,"difficulte":-40.9,"pente":0.024,"en_une_phrase":"Huit exercices de compréhension de texte en anglais : répondre par oui ou non, trouver à quoi renvoie un pronom, relier cause et effet.","notation":"Score global qui résume les résultats des huit exercices.","limites":"Test de 2019, vite dépassé par les modèles. Seuls une dizaine d'anciens modèles y figurent.","mainteneur":"Wang et al. (2019)","url":"https://super.gluebenchmark.com/","verifie":true},"tache":{"id":"autre","nom":"Tests historiques","poids_general":0,"benchmarks_actifs":0},"modeles_mesures":8,"mesures":10,"classement":[{"rang":1,"modele":"t5-11b","nom":"T5-11B","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/t5-11b/","score_publie":0.889,"score_0_1":0.889,"niveau_suggere":45.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":2,"modele":"t5-3b","nom":"T5-3B","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/t5-3b/","score_publie":0.864,"score_0_1":0.864,"niveau_suggere":36.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":3,"modele":"switch-large","nom":"Switch-Large","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/switch-large/","score_publie":0.847,"score_0_1":0.847,"niveau_suggere":30.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":4,"modele":"t5-large","nom":"T5-Large","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/t5-large/","score_publie":0.827,"score_0_1":0.827,"niveau_suggere":24.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":5,"modele":"t5-base","nom":"T5-Base","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/t5-base/","score_publie":0.762,"score_0_1":0.762,"niveau_suggere":7.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":6,"modele":"switch-base","nom":"Switch-Base","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/switch-base/","score_publie":0.733,"score_0_1":0.733,"niveau_suggere":1.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":7,"modele":"instructgpt-175b","nom":"InstructGPT 175B","editeur":"OpenAI","page":"https://quelleia.com/modeles/instructgpt-175b/","score_publie":0.718,"score_0_1":0.718,"niveau_suggere":-1.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":8,"modele":"t5-small","nom":"T5-Small","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/t5-small/","score_publie":0.633,"score_0_1":0.633,"niveau_suggere":-18.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"}]}