{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/modeles/claude-3-5-sonnet-october-2024/","edition":"2026-W40","date":"2026-09-28","echelle":"Score IA : 100 pour le meilleur modèle au lancement du site, 50 pour GPT-4o de novembre 2024. bas et haut bornent l'intervalle à 90 %.","licence":"Réutilisation libre en citant « Quelle IA, quelleia.com » et la date de l'édition. Les mesures renvoient à leurs sources d'origine.","modele":{"liens":[{"type":"essayer","libelle":"Application de l'éditeur : Claude","url":"https://claude.ai"},{"type":"api","libelle":"Documentation de l'API (Anthropic)","url":"https://docs.anthropic.com"},{"type":"editeur","libelle":"Site de l'éditeur (Anthropic)","url":"https://www.anthropic.com"}],"id":"claude-3-5-sonnet-october-2024","nom":"Claude 3.5 Sonnet (October 2024)","editeur":"Anthropic","pays":"États-Unis","date_sortie":"2024-10-22","nouveau":false,"poids_ouverts":false,"acces":"API access","licence":null,"local":null,"prix_eur_par_million":null,"hebergeable_ue":null,"classe_energie":null,"contexte_jetons":null,"entrees":null,"niveaux_reflexion":null,"connaissances_jusqu_a":null,"general":{"score":56,"bas":50.4,"haut":61.6,"mesures":13,"poids_couvert":65,"sources":13,"provisoire":false,"historique":false,"rang":164,"groupe":11},"taches":{"raisonnement":{"score":62.8,"bas":55.5,"haut":70.1,"mesures":4,"benchmarks_tache":9,"couverture":44,"provisoire":false,"rang":88},"code":{"score":65.7,"bas":57,"haut":74.4,"mesures":2,"benchmarks_tache":13,"couverture":15,"provisoire":true},"agents":{"score":59.4,"bas":52,"haut":66.9,"mesures":2,"benchmarks_tache":9,"couverture":22,"provisoire":true},"maths":{"score":48.2,"bas":38.2,"haut":58.1,"mesures":3,"benchmarks_tache":8,"couverture":38,"provisoire":true},"sciences":{"score":56,"bas":51,"haut":61.1,"mesures":2,"benchmarks_tache":5,"couverture":40,"provisoire":true}},"reflexions_mesurees":[],"harnais_mesures":["OpenHands"],"mesures":[{"benchmark":"mmlu","tache":"autre","etat":"archive","score_publie":0.873,"score_0_1":0.8307,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":63.6},{"benchmark":"lech_mazur_writing","tache":"ecriture","etat":"archive","score_publie":8.03,"score_0_1":0.803,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":72.2},{"benchmark":"livebench_code","tache":"code","etat":"archive","score_publie":67.13,"score_0_1":0.6713,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":63.9},{"benchmark":"geobench","tache":"vision","etat":"archive","score_publie":0.62,"score_0_1":0.62,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":56.8},{"benchmark":"livebench","tache":"autre","etat":"archive","score_publie":59.03,"score_0_1":0.5903,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":57.2},{"benchmark":"math_level_5","tache":"maths","etat":"archive","score_publie":0.5694864048338368,"score_0_1":0.5695,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2025-01-27","source":"https://epoch.ai/benchmarks","niveau_suggere":51.3},{"benchmark":"livebench_raisonnement","tache":"raisonnement","etat":"archive","score_publie":56.67,"score_0_1":0.5667,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":54.9},{"benchmark":"livebench_maths","tache":"maths","etat":"archive","score_publie":52.28,"score_0_1":0.5228,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":52.5},{"benchmark":"aider_polyglot","tache":"code","etat":"archive","score_publie":51.6,"score_0_1":0.516,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://aider.chat/docs/leaderboards/","niveau_suggere":65.5},{"benchmark":"cadeval","tache":"vision","etat":"archive","score_publie":0.48,"score_0_1":0.48,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":65.3},{"benchmark":"videomme","tache":"vision","etat":"archive","score_publie":0.6,"score_0_1":0.4667,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":34.3},{"benchmark":"dtbench","tache":"raisonnement","etat":"sature","score_publie":0.6784,"score_0_1":0.464,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://conceptualreasoning.ai/dtbench","niveau_suggere":60.1},{"benchmark":"metr_time_horizons","tache":"agents","etat":"actif","score_publie":0.452356,"score_0_1":0.4524,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/","niveau_suggere":57.7},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.553030303030303,"score_0_1":0.404,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2025-01-27","source":"https://epoch.ai/benchmarks","niveau_suggere":56.3},{"benchmark":"weirdml","tache":"code","etat":"actif","score_publie":0.3997,"score_0_1":0.3997,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://htihle.github.io/weirdml.html","niveau_suggere":67.7},{"benchmark":"balrog","tache":"agents","etat":"actif","score_publie":0.326,"score_0_1":0.326,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://balrogai.com/","niveau_suggere":65.5},{"benchmark":"simplebench","tache":"raisonnement","etat":"actif","score_publie":0.414,"score_0_1":0.2968,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":70.1},{"benchmark":"the_agent_company","tache":"agents","etat":"archive","score_publie":0.24,"score_0_1":0.24,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":65},{"benchmark":"forecastbench","tache":"raisonnement","etat":"actif","score_publie":60.7,"score_0_1":0.214,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":83.5},{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.0847222222222222,"score_0_1":0.0838,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2025-02-25","source":"https://epoch.ai/benchmarks","niveau_suggere":46},{"benchmark":"gso_bench","tache":"code","etat":"actif","score_publie":0.046,"score_0_1":0.046,"reflexion":"inconnue","harnais":"OpenHands","provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":71.3},{"benchmark":"frontiermath","tache":"maths","etat":"actif","score_publie":0.0206896551724137,"score_0_1":0.0363,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2025-03-06","source":"https://epoch.ai/benchmarks","niveau_suggere":60.4},{"benchmark":"enigmaeval","tache":"raisonnement","etat":"actif","score_publie":0.0091,"score_0_1":0.0091,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://labs.scale.com/leaderboard/enigma_eval","niveau_suggere":54.4},{"benchmark":"frontiermath_tier_4","tache":"maths","etat":"actif","score_publie":0,"score_0_1":0,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2025-07-01","source":"https://epoch.ai/benchmarks","niveau_suggere":66},{"benchmark":"hle","tache":"sciences","etat":"actif","score_publie":0.0408,"score_0_1":0,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":44.5},{"benchmark":"vpct","tache":"vision","etat":"archive","score_publie":0.33,"score_0_1":0,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":62.4}]}}