{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/benchmarks/spatialviz_bench/","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":"spatialviz_bench","nom":"SpatialViz-Bench","tache":"vision","etat":"archive","echelle":"unite","modeles":8,"poids_general":0,"difficulte":108.1,"pente":0.0119,"en_une_phrase":"1 180 questions de visualisation dans l'espace : rotation mentale, pliage, coupe d'objets, mécanismes en mouvement.","notation":"Part de bonnes réponses à choix multiples ; les humains font environ 82 %.","limites":"Huit modèles seulement, recopiés de l'article d'origine de 2025, sans mise à jour depuis.","mainteneur":"Auteurs de l'article SpatialViz-Bench (2025)","url":"https://arxiv.org/abs/2507.07610","verifie":true},"tache":{"id":"vision","nom":"Vision et multimodal","poids_general":5,"benchmarks_actifs":3},"modeles_mesures":8,"mesures":8,"classement":[{"rang":1,"modele":"gemini-2-5-pro-jun-2025","nom":"Gemini 2.5 Pro (Jun 2025)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-2-5-pro-jun-2025/","score_publie":0.4466,"score_0_1":0.4466,"niveau_suggere":90,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://arxiv.org/abs/2507.07610"},{"rang":2,"modele":"o1","nom":"o1","editeur":"OpenAI","page":"https://quelleia.com/modeles/o1/","score_publie":0.4136,"score_0_1":0.4136,"niveau_suggere":78.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://arxiv.org/abs/2507.07610"},{"rang":3,"modele":"gemini-2-5-flash-jun-2025","nom":"Gemini 2.5 Flash (Jun 2025)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-2-5-flash-jun-2025/","score_publie":0.3686,"score_0_1":0.3686,"niveau_suggere":62.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://arxiv.org/abs/2507.07610"},{"rang":4,"modele":"llama-4-scout","nom":"Llama 4 Scout","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-4-scout/","score_publie":0.3424,"score_0_1":0.3424,"niveau_suggere":53.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://arxiv.org/abs/2507.07610"},{"rang":5,"modele":"claude-3-7-sonnet","nom":"Claude 3.7 Sonnet","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-3-7-sonnet/","score_publie":0.3389999999999999,"score_0_1":0.339,"niveau_suggere":51.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://arxiv.org/abs/2507.07610"},{"rang":6,"modele":"qwen2-5-72b","nom":"Qwen2.5-72B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen2-5-72b/","score_publie":0.3331,"score_0_1":0.3331,"niveau_suggere":49.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://arxiv.org/abs/2507.07610"},{"rang":7,"modele":"qwen-vl-max","nom":"Qwen-VL-Max","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen-vl-max/","score_publie":0.3203,"score_0_1":0.3203,"niveau_suggere":44.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://arxiv.org/abs/2507.07610"},{"rang":8,"modele":"llama-4-maverick","nom":"Llama 4 Maverick","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-4-maverick/","score_publie":0.3178,"score_0_1":0.3178,"niveau_suggere":43.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://arxiv.org/abs/2507.07610"}]}