{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/benchmarks/hle/","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":"hle","nom":"Humanity's Last Exam","tache":"sciences","etat":"actif","echelle":"unite","modeles":44,"poids_general":2,"difficulte":95.6,"pente":0.09,"en_une_phrase":"2 500 questions de niveau expert dans plus de cent disciplines, écrites par près de mille chercheurs pour mettre les modèles en échec.","notation":"Part de réponses exactes ; le classement publie aussi l'erreur de calibration, qui mesure l'excès de confiance du modèle.","limites":"Epoch le classe « défectueux » : sur 48 questions vérifiées, 22 avaient un énoncé ou un corrigé fautif, ce qui borne le score atteignable.","mainteneur":"Center for AI Safety et Scale AI","url":"https://lastexam.ai/","verifie":true},"tache":{"id":"sciences","nom":"Sciences et savoir expert","poids_general":10,"benchmarks_actifs":5},"modeles_mesures":44,"mesures":54,"classement":[{"rang":1,"modele":"gpt-6-astra","nom":"GPT-6 Astra","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-6-astra/","score_publie":0.5479999999999999,"score_0_1":0.5252,"niveau_suggere":96.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":2,"modele":"claude-fable-5-1","nom":"Claude Fable 5.1","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-fable-5-1/","score_publie":0.465,"score_0_1":0.438,"niveau_suggere":92.8,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":3,"modele":"gemini-3-1-pro","nom":"Gemini 3.1 Pro","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-1-pro/","score_publie":0.4644,"score_0_1":0.4374,"niveau_suggere":92.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":4,"modele":"gemini-3-8-flash","nom":"Gemini 3.8 Flash","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-8-flash/","score_publie":0.4452,"score_0_1":0.4172,"niveau_suggere":91.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":5,"modele":"gpt-5-4-pro","nom":"GPT-5.4 Pro","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-4-pro/","score_publie":0.4432,"score_0_1":0.4151,"niveau_suggere":91.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":6,"modele":"muse-spark","nom":"Muse Spark","editeur":"Meta AI","page":"https://quelleia.com/modeles/muse-spark/","score_publie":0.4056,"score_0_1":0.3756,"niveau_suggere":89.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":7,"modele":"gemini-3-pro","nom":"Gemini 3 Pro","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-pro/","score_publie":0.3752,"score_0_1":0.3437,"niveau_suggere":88.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":8,"modele":"gpt-5-4","nom":"GPT-5.4","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-4/","score_publie":0.3624,"score_0_1":0.3303,"niveau_suggere":87.7,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":9,"modele":"claude-opus-4-7","nom":"Claude Opus 4.7","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-7/","score_publie":0.362,"score_0_1":0.3298,"niveau_suggere":87.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":10,"modele":"claude-opus-4-6","nom":"Claude Opus 4.6","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-6/","score_publie":0.3444,"score_0_1":0.3113,"niveau_suggere":86.8,"reflexion":"maximale","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":11,"modele":"gpt-5-pro","nom":"GPT-5 Pro","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-pro/","score_publie":0.3164,"score_0_1":0.2819,"niveau_suggere":85.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"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":0.278,"score_0_1":0.2416,"niveau_suggere":82.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":13,"modele":"gpt-5","nom":"GPT-5","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5/","score_publie":0.2532,"score_0_1":0.2155,"niveau_suggere":81.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"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":0.252,"score_0_1":0.2143,"niveau_suggere":81.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":15,"modele":"kimi-k2-5","nom":"Kimi K2.5","editeur":"Moonshot","page":"https://quelleia.com/modeles/kimi-k2-5/","score_publie":0.2437,"score_0_1":0.2056,"niveau_suggere":80.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":16,"modele":"gpt-5-1","nom":"GPT-5.1","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-1/","score_publie":0.2368,"score_0_1":0.1983,"niveau_suggere":80.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":17,"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":0.2164,"score_0_1":0.1769,"niveau_suggere":78.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":18,"modele":"o3","nom":"o3","editeur":"OpenAI","page":"https://quelleia.com/modeles/o3/","score_publie":0.2032,"score_0_1":0.163,"niveau_suggere":77.4,"reflexion":"elevee","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":19,"modele":"gpt-5-mini","nom":"GPT-5 mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-mini/","score_publie":0.1944,"score_0_1":0.1538,"niveau_suggere":76.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":20,"modele":"gemini-2-5-pro-mar-2025","nom":"Gemini 2.5 Pro (Mar 2025)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-2-5-pro-mar-2025/","score_publie":0.1816,"score_0_1":0.1403,"niveau_suggere":75.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":21,"modele":"o4-mini","nom":"o4-mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/o4-mini/","score_publie":0.1808,"score_0_1":0.1395,"niveau_suggere":75.4,"reflexion":"elevee","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":22,"modele":"gemini-2-5-pro-may-2025","nom":"Gemini 2.5 Pro (May 2025)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-2-5-pro-may-2025/","score_publie":0.178,"score_0_1":0.1366,"niveau_suggere":75.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":23,"modele":"claude-sonnet-4-5","nom":"Claude Sonnet 4.5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-sonnet-4-5/","score_publie":0.1372,"score_0_1":0.0937,"niveau_suggere":70.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":24,"modele":"gemini-2-5-flash-apr-2025","nom":"Gemini 2.5 Flash (Apr 2025)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-2-5-flash-apr-2025/","score_publie":0.1208,"score_0_1":0.0765,"niveau_suggere":67.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":25,"modele":"claude-opus-4-1","nom":"Claude Opus 4.1","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-1/","score_publie":0.1152,"score_0_1":0.0706,"niveau_suggere":66.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":26,"modele":"gemini-2-5-flash-may-2025","nom":"Gemini 2.5 Flash (May 2025)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-2-5-flash-may-2025/","score_publie":0.1096,"score_0_1":0.0647,"niveau_suggere":65.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":27,"modele":"claude-opus-4","nom":"Claude Opus 4","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4/","score_publie":0.1072,"score_0_1":0.0622,"niveau_suggere":65.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":28,"modele":"gemini-3-1-flash-lite","nom":"Gemini 3.1 Flash-Lite","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-1-flash-lite/","score_publie":0.0864,"score_0_1":0.0403,"niveau_suggere":60.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":29,"modele":"glm-4-5","nom":"GLM-4.5","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/glm-4-5/","score_publie":0.0832,"score_0_1":0.037,"niveau_suggere":59.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":30,"modele":"o1-pro","nom":"o1-pro","editeur":"OpenAI","page":"https://quelleia.com/modeles/o1-pro/","score_publie":0.0812,"score_0_1":0.0349,"niveau_suggere":58.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":30,"modele":"glm-4-5-air","nom":"GLM-4.5-Air","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/glm-4-5-air/","score_publie":0.0812,"score_0_1":0.0349,"niveau_suggere":58.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":32,"modele":"claude-3-7-sonnet","nom":"Claude 3.7 Sonnet","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-3-7-sonnet/","score_publie":0.0804,"score_0_1":0.034,"niveau_suggere":58.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":33,"modele":"o1","nom":"o1","editeur":"OpenAI","page":"https://quelleia.com/modeles/o1/","score_publie":0.0796,"score_0_1":0.0332,"niveau_suggere":58.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":34,"modele":"claude-sonnet-4","nom":"Claude Sonnet 4","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-sonnet-4/","score_publie":0.0776,"score_0_1":0.0311,"niveau_suggere":57.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":35,"modele":"gemini-2-0-flash-thinking-jan-2025","nom":"Gemini 2.0 Flash Thinking (Jan 2025)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-2-0-flash-thinking-jan-2025/","score_publie":0.0655999999999999,"score_0_1":0.0185,"niveau_suggere":51.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":36,"modele":"llama-4-maverick","nom":"Llama 4 Maverick","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-4-maverick/","score_publie":0.0567999999999999,"score_0_1":0.0092,"niveau_suggere":44.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":37,"modele":"gpt-4-5","nom":"GPT-4.5","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4-5/","score_publie":0.0544,"score_0_1":0.0067,"niveau_suggere":44.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":38,"modele":"gpt-4-1","nom":"GPT-4.1","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4-1/","score_publie":0.054,"score_0_1":0.0063,"niveau_suggere":44.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":39,"modele":"gemini-1-5-pro-sept-2024","nom":"Gemini 1.5 Pro (Sept 2024)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-1-5-pro-sept-2024/","score_publie":0.046,"score_0_1":0,"niveau_suggere":44.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":40,"modele":"mistral-medium-3","nom":"Mistral Medium 3","editeur":"Mistral AI","page":"https://quelleia.com/modeles/mistral-medium-3/","score_publie":0.0452,"score_0_1":0,"niveau_suggere":44.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":41,"modele":"amazon-nova-pro","nom":"Amazon Nova Pro","editeur":"Amazon","page":"https://quelleia.com/modeles/amazon-nova-pro/","score_publie":0.044,"score_0_1":0,"niveau_suggere":44.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":42,"modele":"claude-3-5-sonnet-october-2024","nom":"Claude 3.5 Sonnet (October 2024)","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-3-5-sonnet-october-2024/","score_publie":0.0408,"score_0_1":0,"niveau_suggere":44.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":43,"modele":"amazon-nova-lite","nom":"Amazon Nova Lite","editeur":"Amazon","page":"https://quelleia.com/modeles/amazon-nova-lite/","score_publie":0.0364,"score_0_1":0,"niveau_suggere":44.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":44,"modele":"gpt-4o-nov-2024","nom":"GPT-4o (Nov 2024)","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4o-nov-2024/","score_publie":0.0272,"score_0_1":0,"niveau_suggere":44.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"}]}