{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/benchmarks/lech_mazur_writing/","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":"lech_mazur_writing","nom":"Lech Mazur Writing","tache":"ecriture","etat":"archive","echelle":"autre","modeles":44,"poids_general":0,"difficulte":19.7,"pente":0.0268,"en_une_phrase":"Écrire de courtes nouvelles de 400 à 500 mots qui intègrent dix éléments imposés : personnage, objet, lieu, époque, ton.","notation":"Note moyenne sur 10 donnée par sept modèles correcteurs selon une grille de 16 questions, sur 500 nouvelles.","limites":"Ce sont des IA qui notent des IA, en anglais. Le classement est figé à sa version d'août 2025, les suivantes ayant changé de barème.","mainteneur":"Lech Mazur (projet personnel)","url":"https://github.com/lechmazur/writing","verifie":true},"tache":{"id":"ecriture","nom":"Écriture","poids_general":5,"benchmarks_actifs":1},"modeles_mesures":44,"mesures":49,"classement":[{"rang":1,"modele":"gpt-5","nom":"GPT-5","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5/","score_publie":8.6,"score_0_1":0.86,"niveau_suggere":87.6,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":2,"modele":"kimi-k2-jul-2025","nom":"Kimi K2 (Jul 2025)","editeur":"Moonshot","page":"https://quelleia.com/modeles/kimi-k2-jul-2025/","score_publie":8.56,"score_0_1":0.856,"niveau_suggere":86.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":3,"modele":"claude-opus-4-1","nom":"Claude Opus 4.1","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-1/","score_publie":8.47,"score_0_1":0.847,"niveau_suggere":83.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":4,"modele":"o3-pro","nom":"o3-pro","editeur":"OpenAI","page":"https://quelleia.com/modeles/o3-pro/","score_publie":8.44,"score_0_1":0.844,"niveau_suggere":82.8,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":5,"modele":"o3","nom":"o3","editeur":"OpenAI","page":"https://quelleia.com/modeles/o3/","score_publie":8.39,"score_0_1":0.839,"niveau_suggere":81.4,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":6,"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":8.38,"score_0_1":0.838,"niveau_suggere":81.2,"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":8.36,"score_0_1":0.836,"niveau_suggere":80.6,"reflexion":"budget","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":8,"modele":"gpt-5-mini","nom":"GPT-5 mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-mini/","score_publie":8.31,"score_0_1":0.831,"niveau_suggere":79.3,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":9,"modele":"qwen3-235b-a22b","nom":"Qwen3-235B-A22B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen3-235b-a22b/","score_publie":8.3,"score_0_1":0.83,"niveau_suggere":79,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":9,"modele":"deepseek-r1","nom":"DeepSeek-R1","editeur":"DeepSeek","page":"https://quelleia.com/modeles/deepseek-r1/","score_publie":8.3,"score_0_1":0.83,"niveau_suggere":79,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":11,"modele":"qwen3-235b-a22b-thinking-jul-2025","nom":"Qwen3-235B-A22B-Thinking (Jul 2025)","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen3-235b-a22b-thinking-jul-2025/","score_publie":8.24,"score_0_1":0.824,"niveau_suggere":77.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":12,"modele":"deepseek-r1-may-2025","nom":"DeepSeek-R1 (May 2025)","editeur":"DeepSeek","page":"https://quelleia.com/modeles/deepseek-r1-may-2025/","score_publie":8.19,"score_0_1":0.819,"niveau_suggere":76.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":13,"modele":"gpt-4o-nov-2024","nom":"GPT-4o (Nov 2024)","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4o-nov-2024/","score_publie":8.18,"score_0_1":0.818,"niveau_suggere":75.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":14,"modele":"claude-sonnet-4","nom":"Claude Sonnet 4","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-sonnet-4/","score_publie":8.14,"score_0_1":0.814,"niveau_suggere":74.9,"reflexion":"budget","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":15,"modele":"claude-3-7-sonnet","nom":"Claude 3.7 Sonnet","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-3-7-sonnet/","score_publie":8.11,"score_0_1":0.811,"niveau_suggere":74.2,"reflexion":"budget","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":16,"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":8.09,"score_0_1":0.809,"niveau_suggere":73.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":17,"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":8.05,"score_0_1":0.805,"niveau_suggere":72.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":18,"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":8.03,"score_0_1":0.803,"niveau_suggere":72.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":19,"modele":"qwq-32b","nom":"QwQ-32B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwq-32b/","score_publie":8.02,"score_0_1":0.802,"niveau_suggere":72,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":20,"modele":"gemma-3-27b","nom":"Gemma 3 27B","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemma-3-27b/","score_publie":7.99,"score_0_1":0.799,"niveau_suggere":71.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":21,"modele":"mistral-medium-3","nom":"Mistral Medium 3","editeur":"Mistral AI","page":"https://quelleia.com/modeles/mistral-medium-3/","score_publie":7.73,"score_0_1":0.773,"niveau_suggere":65.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":22,"modele":"gpt-oss-120b","nom":"gpt-oss-120b","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-oss-120b/","score_publie":7.71,"score_0_1":0.771,"niveau_suggere":65.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":23,"modele":"deepseek-v3-mar-2025","nom":"DeepSeek-V3 (Mar 2025)","editeur":"DeepSeek","page":"https://quelleia.com/modeles/deepseek-v3-mar-2025/","score_publie":7.7,"score_0_1":0.77,"niveau_suggere":64.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":24,"modele":"grok-4","nom":"Grok 4","editeur":"xAI","page":"https://quelleia.com/modeles/grok-4/","score_publie":7.69,"score_0_1":0.769,"niveau_suggere":64.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":25,"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":7.65,"score_0_1":0.765,"niveau_suggere":63.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":26,"modele":"grok-3","nom":"Grok 3","editeur":"xAI","page":"https://quelleia.com/modeles/grok-3/","score_publie":7.64,"score_0_1":0.764,"niveau_suggere":63.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":27,"modele":"gpt-4-5","nom":"GPT-4.5","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4-5/","score_publie":7.56,"score_0_1":0.756,"niveau_suggere":62,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":28,"modele":"qwen3-30b-a3b","nom":"Qwen3-30B-A3B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen3-30b-a3b/","score_publie":7.53,"score_0_1":0.753,"niveau_suggere":61.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":29,"modele":"o4-mini","nom":"o4-mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/o4-mini/","score_publie":7.5,"score_0_1":0.75,"niveau_suggere":60.8,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":30,"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":7.38,"score_0_1":0.738,"niveau_suggere":58.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":31,"modele":"grok-3-mini","nom":"Grok-3 mini","editeur":"xAI","page":"https://quelleia.com/modeles/grok-3-mini/","score_publie":7.35,"score_0_1":0.735,"niveau_suggere":57.9,"reflexion":"faible","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":31,"modele":"claude-3-5-haiku","nom":"Claude 3.5 Haiku","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-3-5-haiku/","score_publie":7.35,"score_0_1":0.735,"niveau_suggere":57.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":33,"modele":"glm-4-5","nom":"GLM-4.5","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/glm-4-5/","score_publie":7.34,"score_0_1":0.734,"niveau_suggere":57.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":34,"modele":"qwen2-5-max","nom":"Qwen2.5-Max","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen2-5-max/","score_publie":7.29,"score_0_1":0.729,"niveau_suggere":56.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":35,"modele":"gemini-2-0-flash-dec-2024","nom":"Gemini 2.0 Flash (Dec 2024)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-2-0-flash-dec-2024/","score_publie":7.15,"score_0_1":0.715,"niveau_suggere":54.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":36,"modele":"o1","nom":"o1","editeur":"OpenAI","page":"https://quelleia.com/modeles/o1/","score_publie":7.02,"score_0_1":0.702,"niveau_suggere":51.8,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":37,"modele":"mistral-large-2-jul-2024","nom":"Mistral Large 2 (Jul 2024)","editeur":"Mistral AI","page":"https://quelleia.com/modeles/mistral-large-2-jul-2024/","score_publie":6.9,"score_0_1":0.69,"niveau_suggere":49.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":38,"modele":"gpt-4o-mini","nom":"GPT-4o mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4o-mini/","score_publie":6.72,"score_0_1":0.672,"niveau_suggere":46.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":39,"modele":"o1-mini","nom":"o1-mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/o1-mini/","score_publie":6.49,"score_0_1":0.649,"niveau_suggere":42.7,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":40,"modele":"grok-2-dec-2024","nom":"Grok-2 (Dec 2024)","editeur":"xAI","page":"https://quelleia.com/modeles/grok-2-dec-2024/","score_publie":6.36,"score_0_1":0.636,"niveau_suggere":40.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":41,"modele":"phi-4","nom":"Phi-4","editeur":"Microsoft","page":"https://quelleia.com/modeles/phi-4/","score_publie":6.26,"score_0_1":0.626,"niveau_suggere":39,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":42,"modele":"llama-4-maverick","nom":"Llama 4 Maverick","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-4-maverick/","score_publie":6.2,"score_0_1":0.62,"niveau_suggere":38,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":43,"modele":"o3-mini","nom":"o3-mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/o3-mini/","score_publie":6.17,"score_0_1":0.617,"niveau_suggere":37.6,"reflexion":"elevee","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":44,"modele":"amazon-nova-pro","nom":"Amazon Nova Pro","editeur":"Amazon","page":"https://quelleia.com/modeles/amazon-nova-pro/","score_publie":6.05,"score_0_1":0.605,"niveau_suggere":35.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"}]}