{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/benchmarks/swe_bench_verified/","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":"swe_bench_verified","nom":"SWE-bench Verified","tache":"code","etat":"actif","echelle":"unite","modeles":32,"poids_general":1.15,"difficulte":62.2,"pente":0.0535,"en_une_phrase":"Corriger de vrais bugs signalés sur GitHub dans des projets Python connus, puis faire passer les tests du projet.","notation":"Part des 484 tickets résolus, c'est-à-dire dont les tests passent, dans le harnais d'Epoch (terminal et éditeur de fichiers).","limites":"Les dépôts sont publics et anciens, donc en partie mémorisés. Epoch estime que 5 à 10 % des tickets sont ambigus, et le harnais d'un éditeur peut donner d'autres chiffres.","mainteneur":"Epoch AI pour l'évaluation ; sélection validée par OpenAI à partir de SWE-bench","url":"https://epoch.ai/benchmarks/swe-bench-verified","verifie":true},"tache":{"id":"code","nom":"Code","poids_general":15,"benchmarks_actifs":13},"modeles_mesures":32,"mesures":35,"classement":[{"rang":1,"modele":"claude-opus-4-7","nom":"Claude Opus 4.7","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-7/","score_publie":0.8347107438016529,"score_0_1":0.8347,"niveau_suggere":92.5,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-04-20","source":"https://epoch.ai/benchmarks"},{"rang":2,"modele":"gpt-5-5","nom":"GPT-5.5","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-5/","score_publie":0.8057851239669421,"score_0_1":0.8058,"niveau_suggere":88.8,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-04-24","source":"https://epoch.ai/benchmarks"},{"rang":3,"modele":"gemini-3-5-flash","nom":"Gemini 3.5 Flash","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-5-flash/","score_publie":0.7933884297520661,"score_0_1":0.7934,"niveau_suggere":87.4,"reflexion":"elevee","harnais":null,"configurations_mesurees":1,"date":"2026-06-01","source":"https://epoch.ai/benchmarks"},{"rang":4,"modele":"claude-opus-4-6","nom":"Claude Opus 4.6","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-6/","score_publie":0.7871900826446281,"score_0_1":0.7872,"niveau_suggere":86.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":"2026-02-18","source":"https://epoch.ai/benchmarks"},{"rang":5,"modele":"glm-5-2","nom":"GLM-5.2","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/glm-5-2/","score_publie":0.787,"score_0_1":0.787,"niveau_suggere":86.6,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-06-25","source":"https://epoch.ai/benchmarks"},{"rang":6,"modele":"deepseek-v4-pro","nom":"DeepSeek-V4-Pro","editeur":"DeepSeek","page":"https://quelleia.com/modeles/deepseek-v4-pro/","score_publie":0.7763975155279503,"score_0_1":0.7764,"niveau_suggere":85.5,"reflexion":"maximale","harnais":null,"configurations_mesurees":1,"date":"2026-06-18","source":"https://epoch.ai/benchmarks"},{"rang":7,"modele":"qwen3-7-max","nom":"Qwen3.7-Max","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen3-7-max/","score_publie":0.7727272727272727,"score_0_1":0.7727,"niveau_suggere":85.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-06-18","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":0.768595041322314,"score_0_1":0.7686,"niveau_suggere":84.7,"reflexion":"elevee","harnais":null,"configurations_mesurees":1,"date":"2026-03-06","source":"https://epoch.ai/benchmarks"},{"rang":9,"modele":"kimi-k2-6","nom":"Kimi K2.6","editeur":"Moonshot","page":"https://quelleia.com/modeles/kimi-k2-6/","score_publie":0.7665289256198347,"score_0_1":0.7665,"niveau_suggere":84.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-05-08","source":"https://epoch.ai/benchmarks"},{"rang":9,"modele":"claude-opus-4-5","nom":"Claude Opus 4.5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-5/","score_publie":0.7665289256198347,"score_0_1":0.7665,"niveau_suggere":84.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-05","source":"https://epoch.ai/benchmarks"},{"rang":9,"modele":"qwen-3-6-max-preview","nom":"Qwen 3.6 Max (Preview)","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen-3-6-max-preview/","score_publie":0.7665289256198347,"score_0_1":0.7665,"niveau_suggere":84.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-05-28","source":"https://epoch.ai/benchmarks"},{"rang":12,"modele":"gemini-3-1-pro","nom":"Gemini 3.1 Pro","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-1-pro/","score_publie":0.756198347107438,"score_0_1":0.7562,"niveau_suggere":83.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-24","source":"https://epoch.ai/benchmarks"},{"rang":13,"modele":"gemini-3-flash","nom":"Gemini 3 Flash","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-flash/","score_publie":0.7541322314049587,"score_0_1":0.7541,"niveau_suggere":83.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-18","source":"https://epoch.ai/benchmarks"},{"rang":14,"modele":"claude-sonnet-4-6","nom":"Claude Sonnet 4.6","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-sonnet-4-6/","score_publie":0.7520661157024794,"score_0_1":0.7521,"niveau_suggere":83,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-21","source":"https://epoch.ai/benchmarks"},{"rang":15,"modele":"gpt-5-3-codex","nom":"GPT-5.3 Codex","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-3-codex/","score_publie":0.7479338842975206,"score_0_1":0.7479,"niveau_suggere":82.5,"reflexion":"elevee","harnais":null,"configurations_mesurees":1,"date":"2026-02-25","source":"https://epoch.ai/benchmarks"},{"rang":16,"modele":"glm-5-1","nom":"GLM-5.1","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/glm-5-1/","score_publie":0.7417,"score_0_1":0.7417,"niveau_suggere":81.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-05-15","source":"https://epoch.ai/benchmarks"},{"rang":17,"modele":"gpt-5-2","nom":"GPT-5.2","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-2/","score_publie":0.737603305785124,"score_0_1":0.7376,"niveau_suggere":81.5,"reflexion":"elevee","harnais":null,"configurations_mesurees":1,"date":"2026-02-12","source":"https://epoch.ai/benchmarks"},{"rang":17,"modele":"kimi-k2-5","nom":"Kimi K2.5","editeur":"Moonshot","page":"https://quelleia.com/modeles/kimi-k2-5/","score_publie":0.737603305785124,"score_0_1":0.7376,"niveau_suggere":81.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-17","source":"https://epoch.ai/benchmarks"},{"rang":19,"modele":"gpt-5","nom":"GPT-5","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5/","score_publie":0.7355371900826446,"score_0_1":0.7355,"niveau_suggere":81.3,"reflexion":"elevee","harnais":null,"configurations_mesurees":2,"date":"2026-02-06","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":0.7334710743801653,"score_0_1":0.7335,"niveau_suggere":81.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-11","source":"https://epoch.ai/benchmarks"},{"rang":21,"modele":"gemini-3-pro","nom":"Gemini 3 Pro","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-pro/","score_publie":0.7293388429752066,"score_0_1":0.7293,"niveau_suggere":80.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-13","source":"https://epoch.ai/benchmarks"},{"rang":22,"modele":"glm-5","nom":"GLM-5","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/glm-5/","score_publie":0.7207792207792207,"score_0_1":0.7208,"niveau_suggere":79.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-15","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":0.7128099173553719,"score_0_1":0.7128,"niveau_suggere":79.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-05","source":"https://epoch.ai/benchmarks"},{"rang":24,"modele":"claude-opus-4","nom":"Claude Opus 4","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4/","score_publie":0.7066115702479339,"score_0_1":0.7066,"niveau_suggere":78.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-06","source":"https://epoch.ai/benchmarks"},{"rang":25,"modele":"gpt-5-1","nom":"GPT-5.1","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-1/","score_publie":0.6797520661157025,"score_0_1":0.6798,"niveau_suggere":76.3,"reflexion":"elevee","harnais":null,"configurations_mesurees":2,"date":"2026-02-18","source":"https://epoch.ai/benchmarks"},{"rang":26,"modele":"gpt-5-mini","nom":"GPT-5 mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-mini/","score_publie":0.6466942148760331,"score_0_1":0.6467,"niveau_suggere":73.5,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":"2026-02-01","source":"https://epoch.ai/benchmarks"},{"rang":27,"modele":"o3","nom":"o3","editeur":"OpenAI","page":"https://quelleia.com/modeles/o3/","score_publie":0.6231884057971014,"score_0_1":0.6232,"niveau_suggere":71.6,"reflexion":"moyenne","harnais":null,"configurations_mesurees":1,"date":"2026-02-12","source":"https://epoch.ai/benchmarks"},{"rang":28,"modele":"claude-3-7-sonnet","nom":"Claude 3.7 Sonnet","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-3-7-sonnet/","score_publie":0.609504132231405,"score_0_1":0.6095,"niveau_suggere":70.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-04","source":"https://epoch.ai/benchmarks"},{"rang":29,"modele":"qwen-3-6-plus","nom":"Qwen 3.6 Plus","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen-3-6-plus/","score_publie":0.5785123966942148,"score_0_1":0.5785,"niveau_suggere":68.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-05-14","source":"https://epoch.ai/benchmarks"},{"rang":30,"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":0.5755693581780539,"score_0_1":0.5756,"niveau_suggere":67.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-13","source":"https://epoch.ai/benchmarks"},{"rang":31,"modele":"gpt-4-1","nom":"GPT-4.1","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4-1/","score_publie":0.4854166666666666,"score_0_1":0.4854,"niveau_suggere":61.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-08","source":"https://epoch.ai/benchmarks"},{"rang":32,"modele":"gpt-4o-nov-2024","nom":"GPT-4o (Nov 2024)","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4o-nov-2024/","score_publie":0.3099173553719008,"score_0_1":0.3099,"niveau_suggere":47.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":"2026-02-11","source":"https://epoch.ai/benchmarks"}]}