{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/benchmarks/bbh/","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":"bbh","nom":"BIG-Bench Hard (BBH)","tache":"autre","etat":"archive","echelle":"unite","modeles":45,"poids_general":0,"difficulte":18.7,"pente":0.0409,"en_une_phrase":"23 exercices de logique et de raisonnement en plusieurs étapes, retenus en 2022 parce que les modèles y échouaient.","notation":"Moyenne des bonnes réponses sur l'ensemble des exercices.","limites":"Jeu public de 2022, largement vu à l'entraînement et saturé. Les scores viennent d'articles aux réglages variés.","mainteneur":"Suzgun et al. (Google Research et Stanford), à partir de BIG-bench","url":"https://github.com/suzgunmirac/BIG-Bench-Hard","verifie":true},"tache":{"id":"autre","nom":"Tests historiques","poids_general":0,"benchmarks_actifs":0},"modeles_mesures":45,"mesures":88,"classement":[{"rang":1,"modele":"gemini-1-5-pro-may-2024","nom":"Gemini 1.5 Pro (May 2024)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-1-5-pro-may-2024/","score_publie":0.892,"score_0_1":0.856,"niveau_suggere":62.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":2,"modele":"deepseek-v3","nom":"DeepSeek-V3","editeur":"DeepSeek","page":"https://quelleia.com/modeles/deepseek-v3/","score_publie":0.875,"score_0_1":0.8333,"niveau_suggere":58.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":3,"modele":"gemini-1-5-pro-feb-2024","nom":"Gemini 1.5 Pro (Feb 2024)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-1-5-pro-feb-2024/","score_publie":0.84,"score_0_1":0.7867,"niveau_suggere":50.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":4,"modele":"llama-3-1-405b","nom":"Llama 3.1-405B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-3-1-405b/","score_publie":0.829,"score_0_1":0.772,"niveau_suggere":48.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":5,"modele":"phi-3-medium-14b","nom":"phi-3-medium 14B","editeur":"Microsoft","page":"https://quelleia.com/modeles/phi-3-medium-14b/","score_publie":0.814,"score_0_1":0.752,"niveau_suggere":45.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":6,"modele":"qwen2-5-72b","nom":"Qwen2.5-72B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen2-5-72b/","score_publie":0.798,"score_0_1":0.7307,"niveau_suggere":43.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":7,"modele":"phi-3-small-7-4b","nom":"phi-3-small 7.4B","editeur":"Microsoft","page":"https://quelleia.com/modeles/phi-3-small-7-4b/","score_publie":0.791,"score_0_1":0.7213,"niveau_suggere":42,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":8,"modele":"deepseek-v2-moe-236b-may-2024","nom":"DeepSeek-V2 (MoE-236B, May 2024)","editeur":"DeepSeek","page":"https://quelleia.com/modeles/deepseek-v2-moe-236b-may-2024/","score_publie":0.788,"score_0_1":0.7173,"niveau_suggere":41.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":9,"modele":"gpt-4-jun-2023","nom":"GPT-4 (Jun 2023)","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4-jun-2023/","score_publie":0.7512,"score_0_1":0.6683,"niveau_suggere":35.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":10,"modele":"yi-34b","nom":"Yi-34B","editeur":"01.AI","page":"https://quelleia.com/modeles/yi-34b/","score_publie":0.717,"score_0_1":0.6227,"niveau_suggere":31,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":10,"modele":"phi-3-mini-3-8b","nom":"phi-3-mini 3.8B","editeur":"Microsoft","page":"https://quelleia.com/modeles/phi-3-mini-3-8b/","score_publie":0.717,"score_0_1":0.6227,"niveau_suggere":31,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":12,"modele":"stable-beluga-2","nom":"Stable Beluga 2","editeur":"Stability AI","page":"https://quelleia.com/modeles/stable-beluga-2/","score_publie":0.693,"score_0_1":0.5907,"niveau_suggere":27.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":13,"modele":"llama-2-70b","nom":"Llama 2-70B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-2-70b/","score_publie":0.649,"score_0_1":0.532,"niveau_suggere":21.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":4,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":14,"modele":"gpt-3-5-turbo-jun-2023","nom":"GPT-3.5 Turbo (Jun 2023)","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-3-5-turbo-jun-2023/","score_publie":0.6159,"score_0_1":0.4879,"niveau_suggere":17.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":15,"modele":"phi-2","nom":"Phi-2","editeur":"Microsoft","page":"https://quelleia.com/modeles/phi-2/","score_publie":0.594,"score_0_1":0.4587,"niveau_suggere":14.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":16,"modele":"nemotron-4-15b","nom":"Nemotron-4 15B","editeur":"NVIDIA","page":"https://quelleia.com/modeles/nemotron-4-15b/","score_publie":0.587,"score_0_1":0.4493,"niveau_suggere":13.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":17,"modele":"llama-65b","nom":"LLaMA-65B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-65b/","score_publie":0.584,"score_0_1":0.4453,"niveau_suggere":13.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":18,"modele":"llama-2-13b","nom":"Llama 2-13B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-2-13b/","score_publie":0.582,"score_0_1":0.4427,"niveau_suggere":13.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":6,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":19,"modele":"mistral-7b-v0-1","nom":"Mistral 7B v0.1","editeur":"Mistral AI","page":"https://quelleia.com/modeles/mistral-7b-v0-1/","score_publie":0.561,"score_0_1":0.4147,"niveau_suggere":10.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":20,"modele":"gemma-7b","nom":"Gemma 7B","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemma-7b/","score_publie":0.551,"score_0_1":0.4013,"niveau_suggere":9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":21,"modele":"qwen-14b","nom":"Qwen-14B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen-14b/","score_publie":0.55,"score_0_1":0.4,"niveau_suggere":8.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":5,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":22,"modele":"internlm-20b","nom":"internlm-20b","editeur":"Shanghai AI Laboratory","page":"https://quelleia.com/modeles/internlm-20b/","score_publie":0.525,"score_0_1":0.3667,"niveau_suggere":5.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":23,"modele":"llama-33b","nom":"LLaMA-33B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-33b/","score_publie":0.5,"score_0_1":0.3333,"niveau_suggere":1.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":24,"modele":"baichuan2-13b","nom":"Baichuan2-13B","editeur":"Baichuan","page":"https://quelleia.com/modeles/baichuan2-13b/","score_publie":0.49,"score_0_1":0.32,"niveau_suggere":0.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":25,"modele":"baichuan2-13b-chat","nom":"Baichuan2-13B-Chat","editeur":"Baichuan AI","page":"https://quelleia.com/modeles/baichuan2-13b-chat/","score_publie":0.472,"score_0_1":0.296,"niveau_suggere":-2.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":25,"modele":"yi-6b","nom":"Yi 6B","editeur":"01.AI","page":"https://quelleia.com/modeles/yi-6b/","score_publie":0.472,"score_0_1":0.296,"niveau_suggere":-2.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":27,"modele":"qwen-7b","nom":"Qwen-7B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen-7b/","score_publie":0.45,"score_0_1":0.2667,"niveau_suggere":-6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":28,"modele":"llama-2-34b","nom":"Llama 2-34B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-2-34b/","score_publie":0.441,"score_0_1":0.2547,"niveau_suggere":-7.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":29,"modele":"vicuna-13b-v1-1","nom":"vicuna-13b-v1.1","editeur":null,"page":"https://quelleia.com/modeles/vicuna-13b-v1-1/","score_publie":0.4304,"score_0_1":0.2405,"niveau_suggere":-9.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":30,"modele":"baichuan-1-13b","nom":"Baichuan 1-13B","editeur":"Baichuan","page":"https://quelleia.com/modeles/baichuan-1-13b/","score_publie":0.4301,"score_0_1":0.2401,"niveau_suggere":-9.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":31,"modele":"internlm-chat-20b","nom":"internlm-chat-20b","editeur":"Shanghai AI Laboratory","page":"https://quelleia.com/modeles/internlm-chat-20b/","score_publie":0.424,"score_0_1":0.232,"niveau_suggere":-10.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":32,"modele":"baichuan-2-7b","nom":"Baichuan 2-7B","editeur":"Baichuan","page":"https://quelleia.com/modeles/baichuan-2-7b/","score_publie":0.416,"score_0_1":0.2213,"niveau_suggere":-12,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":33,"modele":"llama-2-7b","nom":"Llama 2-7B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-2-7b/","score_publie":0.3916,"score_0_1":0.1888,"niveau_suggere":-16.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":34,"modele":"mpt-30b","nom":"MPT-30B","editeur":"MosaicML","page":"https://quelleia.com/modeles/mpt-30b/","score_publie":0.38,"score_0_1":0.1733,"niveau_suggere":-19.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":35,"modele":"llama-13b","nom":"LLaMA-13B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-13b/","score_publie":0.379,"score_0_1":0.172,"niveau_suggere":-19.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":36,"modele":"falcon-40b","nom":"Falcon-40B","editeur":"Technology Innovation Institute","page":"https://quelleia.com/modeles/falcon-40b/","score_publie":0.371,"score_0_1":0.1613,"niveau_suggere":-21.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":37,"modele":"internlm-7b","nom":"internlm-7b","editeur":"Shanghai AI Laboratory","page":"https://quelleia.com/modeles/internlm-7b/","score_publie":0.37,"score_0_1":0.16,"niveau_suggere":-21.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":38,"modele":"mpt-7b","nom":"MPT-7B","editeur":"MosaicML","page":"https://quelleia.com/modeles/mpt-7b/","score_publie":0.356,"score_0_1":0.1413,"niveau_suggere":-25.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":39,"modele":"gemma-2b","nom":"Gemma 2B","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemma-2b/","score_publie":0.352,"score_0_1":0.136,"niveau_suggere":-26.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":40,"modele":"intellect-1","nom":"INTELLECT-1","editeur":"Prime Intellect","page":"https://quelleia.com/modeles/intellect-1/","score_publie":0.3485,"score_0_1":0.1313,"niveau_suggere":-27.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":41,"modele":"chatglm2-6b","nom":"chatglm2-6b","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/chatglm2-6b/","score_publie":0.337,"score_0_1":0.116,"niveau_suggere":-30.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":42,"modele":"llama-7b","nom":"LLaMA-7B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-7b/","score_publie":0.335,"score_0_1":0.1133,"niveau_suggere":-31.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":43,"modele":"baichuan1-7b","nom":"Baichuan1-7B","editeur":"Baichuan","page":"https://quelleia.com/modeles/baichuan1-7b/","score_publie":0.3248,"score_0_1":0.0997,"niveau_suggere":-35.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":44,"modele":"falcon-7b","nom":"Falcon-7B","editeur":"Technology Innovation Institute","page":"https://quelleia.com/modeles/falcon-7b/","score_publie":0.2877,"score_0_1":0.0503,"niveau_suggere":-53.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":45,"modele":"qwen-1-8b","nom":"Qwen-1_8B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen-1-8b/","score_publie":0.282,"score_0_1":0.0427,"niveau_suggere":-57.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"}]}