{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/benchmarks/ebr_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":"ebr_bench","nom":"EBR-bench (Earthborne Rangers)","tache":"agents","etat":"actif","echelle":"unite","modeles":23,"poids_general":1.67,"difficulte":95.3,"pente":0.1895,"en_une_phrase":"Le modèle rejoue dix fois un jeu de société complexe, Earthborne Rangers, pour voir s'il s'améliore avec l'expérience.","notation":"Part des 21 objectifs atteints lors des deux dernières parties, après huit parties d'apprentissage ; la meilleure des deux est retenue.","limites":"Seuls les cinq premiers jours de la campagne sont joués, sur 20 à 25. Le résultat dépend du harnais et de la façon dont l'historique des parties est résumé.","mainteneur":"Epoch AI","url":"https://epoch.ai/benchmarks/ebr-bench","verifie":true},"tache":{"id":"agents","nom":"Agents","poids_general":15,"benchmarks_actifs":9},"modeles_mesures":23,"mesures":23,"classement":[{"rang":1,"modele":"gpt-6-astra","nom":"GPT-6 Astra","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-6-astra/","score_publie":0.7619047619047619,"score_0_1":0.7619,"niveau_suggere":101.4,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-08-30","source":"https://epoch.ai/benchmarks"},{"rang":2,"modele":"claude-opus-5-5","nom":"Claude Opus 5.5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-5-5/","score_publie":0.7142857142857142,"score_0_1":0.7143,"niveau_suggere":100.1,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-09-22","source":"https://epoch.ai/benchmarks"},{"rang":3,"modele":"claude-fable-5-1","nom":"Claude Fable 5.1","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-fable-5-1/","score_publie":0.5714285714285714,"score_0_1":0.5714,"niveau_suggere":96.8,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-09-08","source":"https://epoch.ai/benchmarks"},{"rang":4,"modele":"gpt-6-sol","nom":"GPT-6 Sol","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-6-sol/","score_publie":0.5333333333333333,"score_0_1":0.5333,"niveau_suggere":96,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-09-22","source":"https://epoch.ai/benchmarks"},{"rang":5,"modele":"claude-opus-5","nom":"Claude Opus 5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-5/","score_publie":0.4571428571428571,"score_0_1":0.4571,"niveau_suggere":94.4,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-09-05","source":"https://epoch.ai/benchmarks"},{"rang":6,"modele":"gpt-5-6-sol","nom":"GPT-5.6 Sol","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-6-sol/","score_publie":0.4476190476190475,"score_0_1":0.4476,"niveau_suggere":94.2,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-09-04","source":"https://epoch.ai/benchmarks"},{"rang":7,"modele":"claude-fable-5","nom":"Claude Fable 5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-fable-5/","score_publie":0.395238,"score_0_1":0.3952,"niveau_suggere":93.1,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-07-17","source":"https://epoch.ai/benchmarks"},{"rang":8,"modele":"gpt-5-5","nom":"GPT-5.5","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-5/","score_publie":0.3428571428571428,"score_0_1":0.3429,"niveau_suggere":91.9,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-07-27","source":"https://epoch.ai/benchmarks"},{"rang":9,"modele":"grok-4-6","nom":"Grok 4.6","editeur":"xAI","page":"https://quelleia.com/modeles/grok-4-6/","score_publie":0.3047619047619047,"score_0_1":0.3048,"niveau_suggere":90.9,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-08-18","source":"https://epoch.ai/benchmarks"},{"rang":10,"modele":"claude-opus-4-8","nom":"Claude Opus 4.8","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-8/","score_publie":0.2857142857142857,"score_0_1":0.2857,"niveau_suggere":90.5,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-08-07","source":"https://epoch.ai/benchmarks"},{"rang":11,"modele":"gpt-5-4","nom":"GPT-5.4","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-4/","score_publie":0.253968,"score_0_1":0.254,"niveau_suggere":89.6,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-06-25","source":"https://epoch.ai/benchmarks"},{"rang":12,"modele":"gpt-5-2","nom":"GPT-5.2","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-2/","score_publie":0.230159,"score_0_1":0.2302,"niveau_suggere":88.9,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-06-26","source":"https://epoch.ai/benchmarks"},{"rang":13,"modele":"claude-opus-4-7","nom":"Claude Opus 4.7","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-7/","score_publie":0.1904761904761904,"score_0_1":0.1905,"niveau_suggere":87.7,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-06-30","source":"https://epoch.ai/benchmarks"},{"rang":14,"modele":"gemini-3-1-pro","nom":"Gemini 3.1 Pro","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-1-pro/","score_publie":0.1428571428571428,"score_0_1":0.1429,"niveau_suggere":85.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-06-25","source":"https://epoch.ai/benchmarks"},{"rang":14,"modele":"claude-opus-4-5","nom":"Claude Opus 4.5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-5/","score_publie":0.1428571428571428,"score_0_1":0.1429,"niveau_suggere":85.8,"reflexion":"budget","harnais":null,"configurations_mesurees":1,"date":"2026-06-25","source":"https://epoch.ai/benchmarks"},{"rang":16,"modele":"claude-opus-4-6","nom":"Claude Opus 4.6","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-6/","score_publie":0.126984,"score_0_1":0.127,"niveau_suggere":85.1,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-06-29","source":"https://epoch.ai/benchmarks"},{"rang":16,"modele":"gpt-5","nom":"GPT-5","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5/","score_publie":0.126984,"score_0_1":0.127,"niveau_suggere":85.1,"reflexion":"elevee","harnais":null,"configurations_mesurees":1,"date":"2026-06-29","source":"https://epoch.ai/benchmarks"},{"rang":18,"modele":"qwen3-7-max","nom":"Qwen3.7-Max","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen3-7-max/","score_publie":0.0952380952380952,"score_0_1":0.0952,"niveau_suggere":83.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-06-25","source":"https://epoch.ai/benchmarks"},{"rang":18,"modele":"glm-5-2","nom":"GLM-5.2","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/glm-5-2/","score_publie":0.0952380952380952,"score_0_1":0.0952,"niveau_suggere":83.4,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-06-29","source":"https://epoch.ai/benchmarks"},{"rang":20,"modele":"claude-opus-4-1","nom":"Claude Opus 4.1","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-1/","score_publie":0.079365,"score_0_1":0.0794,"niveau_suggere":82.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-06-25","source":"https://epoch.ai/benchmarks"},{"rang":21,"modele":"gemini-3-5-flash","nom":"Gemini 3.5 Flash","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-5-flash/","score_publie":0.0476190476190476,"score_0_1":0.0476,"niveau_suggere":79.5,"reflexion":"elevee","harnais":null,"configurations_mesurees":1,"date":"2026-06-25","source":"https://epoch.ai/benchmarks"},{"rang":22,"modele":"kimi-k2-6","nom":"Kimi K2.6","editeur":"Moonshot","page":"https://quelleia.com/modeles/kimi-k2-6/","score_publie":0.0238095238095238,"score_0_1":0.0238,"niveau_suggere":75.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-06-25","source":"https://epoch.ai/benchmarks"},{"rang":22,"modele":"claude-sonnet-4-5","nom":"Claude Sonnet 4.5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-sonnet-4-5/","score_publie":0.0238095238095238,"score_0_1":0.0238,"niveau_suggere":75.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-06-25","source":"https://epoch.ai/benchmarks"}]}