{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/benchmarks/deepswe/","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":"deepswe","nom":"DeepSWE","tache":"code","etat":"actif","echelle":"unite","modeles":26,"poids_general":1.15,"difficulte":86.5,"pente":0.1313,"en_une_phrase":"113 tâches de développement longues et inédites, dans 91 dépôts open source actifs et cinq langages.","notation":"Part d'essais réussis selon un vérificateur écrit à la main, sur quatre passages avec le harnais mini-SWE-agent.","limites":"Epoch le classe « défectueux » : au moins 23 tâches sur 113 échouent à cause du vérificateur lui-même, ce qui pénalise des solutions correctes.","mainteneur":"Datacurve","url":"https://deepswe.datacurve.ai/","verifie":true},"tache":{"id":"code","nom":"Code","poids_general":15,"benchmarks_actifs":13},"modeles_mesures":26,"mesures":69,"classement":[{"rang":1,"modele":"gpt-6-astra","nom":"GPT-6 Astra","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-6-astra/","score_publie":0.7411504424778761,"score_0_1":0.7412,"niveau_suggere":94.5,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":5,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":2,"modele":"gemini-3-8-flash","nom":"Gemini 3.8 Flash","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-8-flash/","score_publie":0.738255033557047,"score_0_1":0.7383,"niveau_suggere":94.4,"reflexion":"elevee","harnais":"mini-SWE-agent","configurations_mesurees":2,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":3,"modele":"claude-opus-5","nom":"Claude Opus 5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-5/","score_publie":0.7364864864864865,"score_0_1":0.7365,"niveau_suggere":94.3,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":5,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":4,"modele":"gpt-5-6-sol","nom":"GPT-5.6 Sol","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-6-sol/","score_publie":0.7266666666666667,"score_0_1":0.7267,"niveau_suggere":94,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":5,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":5,"modele":"claude-fable-5","nom":"Claude Fable 5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-fable-5/","score_publie":0.6991150442477876,"score_0_1":0.6991,"niveau_suggere":92.9,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":5,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":6,"modele":"gpt-5-6-terra","nom":"GPT-5.6 Terra","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-6-terra/","score_publie":0.6962305986696231,"score_0_1":0.6962,"niveau_suggere":92.8,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":5,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":7,"modele":"glm-5-3","nom":"GLM-5.3","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/glm-5-3/","score_publie":0.6895787139689579,"score_0_1":0.6896,"niveau_suggere":92.6,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":8,"modele":"kimi-k3","nom":"Kimi K3","editeur":"Moonshot","page":"https://quelleia.com/modeles/kimi-k3/","score_publie":0.6851441241685144,"score_0_1":0.6851,"niveau_suggere":92.4,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":9,"modele":"grok-4-6","nom":"Grok 4.6","editeur":"xAI","page":"https://quelleia.com/modeles/grok-4-6/","score_publie":0.6747787610619469,"score_0_1":0.6748,"niveau_suggere":92.1,"reflexion":"moyenne","harnais":"mini-SWE-agent","configurations_mesurees":4,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":10,"modele":"gpt-5-6-luna","nom":"GPT-5.6 Luna","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-6-luna/","score_publie":0.671875,"score_0_1":0.6719,"niveau_suggere":92,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":5,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":11,"modele":"gpt-5-5","nom":"GPT-5.5","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-5/","score_publie":0.6703539823008849,"score_0_1":0.6704,"niveau_suggere":91.9,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":4,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":12,"modele":"gemini-3-7-flash","nom":"Gemini 3.7 Flash","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-7-flash/","score_publie":0.6548672566371682,"score_0_1":0.6549,"niveau_suggere":91.4,"reflexion":"moyenne","harnais":"mini-SWE-agent","configurations_mesurees":3,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":13,"modele":"glm-5-3-flash","nom":"GLM-5.3-Flash","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/glm-5-3-flash/","score_publie":0.6339285714285714,"score_0_1":0.6339,"niveau_suggere":90.7,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":14,"modele":"claude-opus-4-8","nom":"Claude Opus 4.8","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-opus-4-8/","score_publie":0.5897435897435898,"score_0_1":0.5897,"niveau_suggere":89.3,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":5,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":15,"modele":"qwen-3-8-max","nom":"Qwen 3.8 Max","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen-3-8-max/","score_publie":0.5746102449888641,"score_0_1":0.5746,"niveau_suggere":88.8,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":16,"modele":"muse-spark-1-2","nom":"Muse Spark 1.2","editeur":"Meta AI","page":"https://quelleia.com/modeles/muse-spark-1-2/","score_publie":0.5486725663716814,"score_0_1":0.5487,"niveau_suggere":88,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":17,"modele":"claude-sonnet-5","nom":"Claude Sonnet 5","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-sonnet-5/","score_publie":0.5384615384615384,"score_0_1":0.5385,"niveau_suggere":87.7,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":5,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":18,"modele":"grok-4-5","nom":"Grok 4.5","editeur":"xAI","page":"https://quelleia.com/modeles/grok-4-5/","score_publie":0.5376106194690266,"score_0_1":0.5376,"niveau_suggere":87.7,"reflexion":"elevee","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":19,"modele":"muse-spark-1-1","nom":"Muse Spark 1.1","editeur":"Meta AI","page":"https://quelleia.com/modeles/muse-spark-1-1/","score_publie":0.5331858407079646,"score_0_1":0.5332,"niveau_suggere":87.5,"reflexion":"inconnue","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":20,"modele":"gpt-5-4","nom":"GPT-5.4","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-5-4/","score_publie":0.5176991150442478,"score_0_1":0.5177,"niveau_suggere":87,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":21,"modele":"gemini-3-6-flash","nom":"Gemini 3.6 Flash","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-6-flash/","score_publie":0.4668141592920354,"score_0_1":0.4668,"niveau_suggere":85.5,"reflexion":"elevee","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":22,"modele":"glm-5-2","nom":"GLM-5.2","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/glm-5-2/","score_publie":0.4377777777777778,"score_0_1":0.4378,"niveau_suggere":84.6,"reflexion":"maximale","harnais":"mini-SWE-agent","configurations_mesurees":2,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":23,"modele":"gemini-3-5-flash","nom":"Gemini 3.5 Flash","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-5-flash/","score_publie":0.3738938053097345,"score_0_1":0.3739,"niveau_suggere":82.6,"reflexion":"moyenne","harnais":"mini-SWE-agent","configurations_mesurees":2,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":24,"modele":"kimi-k2-7-code","nom":"Kimi K2.7 Code","editeur":"Moonshot","page":"https://quelleia.com/modeles/kimi-k2-7-code/","score_publie":0.3053097345132743,"score_0_1":0.3053,"niveau_suggere":80.3,"reflexion":"inconnue","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":25,"modele":"claude-sonnet-4-6","nom":"Claude Sonnet 4.6","editeur":"Anthropic","page":"https://quelleia.com/modeles/claude-sonnet-4-6/","score_publie":0.2993348115299334,"score_0_1":0.2993,"niveau_suggere":80,"reflexion":"elevee","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"},{"rang":26,"modele":"gemini-3-1-pro","nom":"Gemini 3.1 Pro","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-3-1-pro/","score_publie":0.1172566371681416,"score_0_1":0.1173,"niveau_suggere":71.1,"reflexion":"inconnue","harnais":"mini-SWE-agent","configurations_mesurees":1,"date":null,"source":"https://deepswe.datacurve.ai/"}]}