{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/benchmarks/cybench/","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":"cybench","nom":"Cybench","tache":"securite","etat":"archive","echelle":"unite","modeles":21,"poids_general":0,"difficulte":74.7,"pente":0.1565,"en_une_phrase":"40 épreuves de piratage tirées de compétitions professionnelles : cryptographie, failles web, rétro-ingénierie, analyse de traces.","notation":"Part d'épreuves résolues sans aucune indication intermédiaire (mode « unguided »).","limites":"Une partie des chiffres vient des fiches des éditeurs, avec leurs propres harnais. Le meilleur modèle dépasse 90 % et les épreuves sont publiques.","mainteneur":"Université Stanford","url":"https://cybench.github.io/","verifie":true},"tache":{"id":"securite","nom":"Sécurité","poids_general":0,"benchmarks_actifs":0},"modeles_mesures":21,"mesures":22,"classement":[{"rang":1,"modele":"claude-opus-4-6","nom":"Claude Opus 4.6","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-6/","score_publie":0.93,"score_0_1":0.93,"niveau_suggere":91.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":2,"modele":"claude-opus-4-5","nom":"Claude Opus 4.5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-5/","score_publie":0.82,"score_0_1":0.82,"niveau_suggere":84.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":3,"modele":"claude-sonnet-4-5","nom":"Claude Sonnet 4.5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-sonnet-4-5/","score_publie":0.6,"score_0_1":0.6,"niveau_suggere":77.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":4,"modele":"grok-4","nom":"Grok 4","editeur":"xAI","page":"https://quelleia.com/modeles/grok-4/","score_publie":0.43,"score_0_1":0.43,"niveau_suggere":72.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":5,"modele":"claude-opus-4-1","nom":"Claude Opus 4.1","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-1/","score_publie":0.42,"score_0_1":0.42,"niveau_suggere":72.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":6,"modele":"grok-4-1","nom":"Grok 4.1","editeur":"xAI","page":"https://quelleia.com/modeles/grok-4-1/","score_publie":0.39,"score_0_1":0.39,"niveau_suggere":71.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":7,"modele":"claude-opus-4","nom":"Claude Opus 4","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4/","score_publie":0.38,"score_0_1":0.38,"niveau_suggere":71.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":8,"modele":"claude-sonnet-4","nom":"Claude Sonnet 4","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-sonnet-4/","score_publie":0.35,"score_0_1":0.35,"niveau_suggere":70.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":9,"modele":"grok-4-fast","nom":"Grok 4 Fast","editeur":"xAI","page":"https://quelleia.com/modeles/grok-4-fast/","score_publie":0.3,"score_0_1":0.3,"niveau_suggere":69.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":10,"modele":"o3-mini","nom":"o3-mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/o3-mini/","score_publie":0.225,"score_0_1":0.225,"niveau_suggere":66.8,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":11,"modele":"claude-3-7-sonnet","nom":"Claude 3.7 Sonnet","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-3-7-sonnet/","score_publie":0.2,"score_0_1":0.2,"niveau_suggere":65.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":12,"modele":"gpt-4-5","nom":"GPT-4.5","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4-5/","score_publie":0.175,"score_0_1":0.175,"niveau_suggere":64.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":12,"modele":"claude-3-5-sonnet","nom":"Claude 3.5 Sonnet","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-3-5-sonnet/","score_publie":0.175,"score_0_1":0.175,"niveau_suggere":64.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":14,"modele":"gpt-4o-nov-2024","nom":"GPT-4o (Nov 2024)","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4o-nov-2024/","score_publie":0.125,"score_0_1":0.125,"niveau_suggere":62.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":15,"modele":"o1-preview","nom":"o1-preview","editeur":"OpenAI","page":"https://quelleia.com/modeles/o1-preview/","score_publie":0.1,"score_0_1":0.1,"niveau_suggere":60.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":15,"modele":"o1-mini","nom":"o1-mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/o1-mini/","score_publie":0.1,"score_0_1":0.1,"niveau_suggere":60.7,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":15,"modele":"claude-3-opus","nom":"Claude 3 Opus","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-3-opus/","score_publie":0.1,"score_0_1":0.1,"niveau_suggere":60.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":18,"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":0.075,"score_0_1":0.075,"niveau_suggere":58.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":18,"modele":"llama-3-1-405b","nom":"Llama 3.1-405B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-3-1-405b/","score_publie":0.075,"score_0_1":0.075,"niveau_suggere":58.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":18,"modele":"mixtral-8x22b","nom":"Mixtral 8x22B","editeur":"Mistral AI","page":"https://quelleia.com/modeles/mixtral-8x22b/","score_publie":0.075,"score_0_1":0.075,"niveau_suggere":58.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":21,"modele":"llama-3-70b","nom":"Llama 3-70B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-3-70b/","score_publie":0.05,"score_0_1":0.05,"niveau_suggere":55.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"}]}