{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/benchmarks/piqa/","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":"piqa","nom":"PIQA","tache":"autre","etat":"archive","echelle":"unite","modeles":59,"poids_general":0,"difficulte":-53.5,"pente":0.0087,"en_une_phrase":"Choisir, entre deux façons de faire, celle qui marche dans le monde physique pour un problème de tous les jours.","notation":"Part de bonnes réponses entre deux choix ; le hasard donne 50 %.","limites":"Test de 2019 saturé. Avec deux choix seulement, le hasard pèse lourd et les écarts entre modèles sont faibles.","mainteneur":"Bisk et al. (2019)","url":"https://arxiv.org/abs/1911.11641","verifie":true},"tache":{"id":"autre","nom":"Tests historiques","poids_general":0,"benchmarks_actifs":0},"modeles_mesures":59,"mesures":113,"classement":[{"rang":1,"modele":"gpt-4o-mini","nom":"GPT-4o mini","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-4o-mini/","score_publie":0.887,"score_0_1":0.774,"niveau_suggere":88.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":2,"modele":"phi-3-5-moe","nom":"Phi-3.5-MoE","editeur":"Microsoft","page":"https://quelleia.com/modeles/phi-3-5-moe/","score_publie":0.886,"score_0_1":0.772,"niveau_suggere":87.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":3,"modele":"gemini-1-5-flash-sep-2024","nom":"Gemini 1.5 Flash (Sep 2024)","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemini-1-5-flash-sep-2024/","score_publie":0.875,"score_0_1":0.75,"niveau_suggere":73.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":4,"modele":"llama-3-1-405b","nom":"Llama 3.1-405B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-3-1-405b/","score_publie":0.859,"score_0_1":0.718,"niveau_suggere":54.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":5,"modele":"falcon-180b","nom":"Falcon-180B","editeur":"Technology Innovation Institute","page":"https://quelleia.com/modeles/falcon-180b/","score_publie":0.849,"score_0_1":0.698,"niveau_suggere":43.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":6,"modele":"deepseek-v3","nom":"DeepSeek-V3","editeur":"DeepSeek","page":"https://quelleia.com/modeles/deepseek-v3/","score_publie":0.847,"score_0_1":0.694,"niveau_suggere":41.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":7,"modele":"inflection-1","nom":"Inflection-1","editeur":"Inflection AI","page":"https://quelleia.com/modeles/inflection-1/","score_publie":0.842,"score_0_1":0.684,"niveau_suggere":35.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":8,"modele":"deepseek-v2-moe-236b-may-2024","nom":"DeepSeek-V2 (MoE-236B, May 2024)","editeur":"DeepSeek","page":"https://quelleia.com/modeles/deepseek-v2-moe-236b-may-2024/","score_publie":0.839,"score_0_1":0.678,"niveau_suggere":32.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":9,"modele":"gemma-2-9b","nom":"Gemma 2 9B","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemma-2-9b/","score_publie":0.837,"score_0_1":0.674,"niveau_suggere":30.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":10,"modele":"mixtral-8x7b","nom":"Mixtral 8x7B","editeur":"Mistral AI","page":"https://quelleia.com/modeles/mixtral-8x7b/","score_publie":0.836,"score_0_1":0.672,"niveau_suggere":29.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":11,"modele":"mistral-nemo","nom":"Mistral NeMo","editeur":"Mistral AI","page":"https://quelleia.com/modeles/mistral-nemo/","score_publie":0.835,"score_0_1":0.67,"niveau_suggere":28.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":12,"modele":"stable-beluga-2","nom":"Stable Beluga 2","editeur":"Stability AI","page":"https://quelleia.com/modeles/stable-beluga-2/","score_publie":0.833,"score_0_1":0.666,"niveau_suggere":26.3,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":13,"modele":"megatron-turing-nlg-530b","nom":"Megatron-Turing NLG 530B","editeur":"Microsoft","page":"https://quelleia.com/modeles/megatron-turing-nlg-530b/","score_publie":0.8319,"score_0_1":0.6638,"niveau_suggere":25.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":4,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":14,"modele":"mistral-7b-v0-1","nom":"Mistral 7B v0.1","editeur":"Mistral AI","page":"https://quelleia.com/modeles/mistral-7b-v0-1/","score_publie":0.83,"score_0_1":0.66,"niveau_suggere":23.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":14,"modele":"falcon-40b","nom":"Falcon-40B","editeur":"Technology Innovation Institute","page":"https://quelleia.com/modeles/falcon-40b/","score_publie":0.83,"score_0_1":0.66,"niveau_suggere":23.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":16,"modele":"llama-2-70b","nom":"Llama 2-70B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-2-70b/","score_publie":0.828,"score_0_1":0.656,"niveau_suggere":21.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":4,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":16,"modele":"llama-65b","nom":"LLaMA-65B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-65b/","score_publie":0.828,"score_0_1":0.656,"niveau_suggere":21.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":18,"modele":"qwen2-5-72b","nom":"Qwen2.5-72B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen2-5-72b/","score_publie":0.826,"score_0_1":0.652,"niveau_suggere":19.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":19,"modele":"nemotron-4-15b","nom":"Nemotron-4 15B","editeur":"NVIDIA","page":"https://quelleia.com/modeles/nemotron-4-15b/","score_publie":0.824,"score_0_1":0.648,"niveau_suggere":17.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":20,"modele":"palm-540b","nom":"PaLM (540B)","editeur":"Google Research","page":"https://quelleia.com/modeles/palm-540b/","score_publie":0.823,"score_0_1":0.646,"niveau_suggere":16.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":20,"modele":"llama-33b","nom":"LLaMA-33B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-33b/","score_publie":0.823,"score_0_1":0.646,"niveau_suggere":16.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":4,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":20,"modele":"instructgpt-175b","nom":"InstructGPT 175B","editeur":"OpenAI","page":"https://quelleia.com/modeles/instructgpt-175b/","score_publie":0.823,"score_0_1":0.646,"niveau_suggere":16.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":8,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":23,"modele":"mistral-7b-v0-2","nom":"Mistral 7B v0.2","editeur":"Mistral AI","page":"https://quelleia.com/modeles/mistral-7b-v0-2/","score_publie":0.822,"score_0_1":0.644,"niveau_suggere":15,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":24,"modele":"mpt-30b","nom":"MPT-30B","editeur":"MosaicML","page":"https://quelleia.com/modeles/mpt-30b/","score_publie":0.819,"score_0_1":0.638,"niveau_suggere":12,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":24,"modele":"llama-2-34b","nom":"Llama 2-34B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-2-34b/","score_publie":0.819,"score_0_1":0.638,"niveau_suggere":12,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":26,"modele":"chinchilla","nom":"Chinchilla","editeur":"DeepMind","page":"https://quelleia.com/modeles/chinchilla/","score_publie":0.818,"score_0_1":0.636,"niveau_suggere":11,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":26,"modele":"gopher-280b","nom":"Gopher (280B)","editeur":"DeepMind","page":"https://quelleia.com/modeles/gopher-280b/","score_publie":0.818,"score_0_1":0.636,"niveau_suggere":11,"reflexion":"inconnue","harnais":null,"configurations_mesurees":4,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":28,"modele":"llama-3-1-8b","nom":"Llama 3.1-8B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-3-1-8b/","score_publie":0.812,"score_0_1":0.624,"niveau_suggere":5.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":28,"modele":"gemma-7b","nom":"Gemma 7B","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemma-7b/","score_publie":0.812,"score_0_1":0.624,"niveau_suggere":5.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":30,"modele":"phi-3-5-mini","nom":"Phi-3.5-mini","editeur":"Microsoft","page":"https://quelleia.com/modeles/phi-3-5-mini/","score_publie":0.81,"score_0_1":0.62,"niveau_suggere":3.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":31,"modele":"llama-2-13b","nom":"Llama 2-13B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-2-13b/","score_publie":0.808,"score_0_1":0.616,"niveau_suggere":1.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":6,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":32,"modele":"mpt-7b","nom":"MPT-7B","editeur":"MosaicML","page":"https://quelleia.com/modeles/mpt-7b/","score_publie":0.806,"score_0_1":0.612,"niveau_suggere":-0.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":4,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":33,"modele":"palm-62b","nom":"PaLM 62B","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/palm-62b/","score_publie":0.805,"score_0_1":0.61,"niveau_suggere":-1.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":34,"modele":"internlm-20b","nom":"internlm-20b","editeur":"Shanghai AI Laboratory","page":"https://quelleia.com/modeles/internlm-20b/","score_publie":0.803,"score_0_1":0.606,"niveau_suggere":-3.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":34,"modele":"falcon-7b","nom":"Falcon-7B","editeur":"Technology Innovation Institute","page":"https://quelleia.com/modeles/falcon-7b/","score_publie":0.803,"score_0_1":0.606,"niveau_suggere":-3.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":5,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":36,"modele":"llama-13b","nom":"LLaMA-13B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-13b/","score_publie":0.801,"score_0_1":0.602,"niveau_suggere":-5.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":3,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":37,"modele":"qwen-14b","nom":"Qwen-14B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen-14b/","score_publie":0.799,"score_0_1":0.598,"niveau_suggere":-7.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":38,"modele":"llama-7b","nom":"LLaMA-7B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-7b/","score_publie":0.798,"score_0_1":0.596,"niveau_suggere":-8.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":5,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":39,"modele":"llama-2-7b","nom":"Llama 2-7B","editeur":"Meta AI","page":"https://quelleia.com/modeles/llama-2-7b/","score_publie":0.788,"score_0_1":0.576,"niveau_suggere":-18.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":6,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":40,"modele":"baichuan2-13b","nom":"Baichuan2-13B","editeur":"Baichuan","page":"https://quelleia.com/modeles/baichuan2-13b/","score_publie":0.781,"score_0_1":0.562,"niveau_suggere":-24.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":2,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":41,"modele":"qwen-7b","nom":"Qwen-7B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen-7b/","score_publie":0.779,"score_0_1":0.558,"niveau_suggere":-26.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":41,"modele":"internlm-7b","nom":"internlm-7b","editeur":"Shanghai AI Laboratory","page":"https://quelleia.com/modeles/internlm-7b/","score_publie":0.779,"score_0_1":0.558,"niveau_suggere":-26.5,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":43,"modele":"vicuna-13b-v1-1","nom":"vicuna-13b-v1.1","editeur":null,"page":"https://quelleia.com/modeles/vicuna-13b-v1-1/","score_publie":0.774,"score_0_1":0.548,"niveau_suggere":-31.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":44,"modele":"gemma-2b","nom":"Gemma 2B","editeur":"Google DeepMind","page":"https://quelleia.com/modeles/gemma-2b/","score_publie":0.773,"score_0_1":0.546,"niveau_suggere":-32.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":45,"modele":"redpajama-incite-7b-base","nom":"RedPajama-INCITE-7B-Base","editeur":null,"page":"https://quelleia.com/modeles/redpajama-incite-7b-base/","score_publie":0.769,"score_0_1":0.538,"niveau_suggere":-35.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":46,"modele":"gpt-neox-20b","nom":"GPT-NeoX-20B","editeur":"EleutherAI","page":"https://quelleia.com/modeles/gpt-neox-20b/","score_publie":0.767,"score_0_1":0.534,"niveau_suggere":-37.7,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":47,"modele":"baichuan1-7b","nom":"Baichuan1-7B","editeur":"Baichuan","page":"https://quelleia.com/modeles/baichuan1-7b/","score_publie":0.762,"score_0_1":0.524,"niveau_suggere":-42.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":48,"modele":"open-llama-7b","nom":"open_llama_7b","editeur":null,"page":"https://quelleia.com/modeles/open-llama-7b/","score_publie":0.76,"score_0_1":0.52,"niveau_suggere":-44.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":49,"modele":"opt-13b","nom":"opt-13b","editeur":"Meta AI","page":"https://quelleia.com/modeles/opt-13b/","score_publie":0.757,"score_0_1":0.514,"niveau_suggere":-47,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":50,"modele":"xgen-7b","nom":"XGen-7B","editeur":"Salesforce","page":"https://quelleia.com/modeles/xgen-7b/","score_publie":0.755,"score_0_1":0.51,"niveau_suggere":-48.9,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":51,"modele":"dolly-2-0-12b","nom":"Dolly 2.0-12b","editeur":"Databricks","page":"https://quelleia.com/modeles/dolly-2-0-12b/","score_publie":0.754,"score_0_1":0.508,"niveau_suggere":-49.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":51,"modele":"gpt-j-6b","nom":"GPT-J-6B","editeur":"EleutherAI","page":"https://quelleia.com/modeles/gpt-j-6b/","score_publie":0.754,"score_0_1":0.508,"niveau_suggere":-49.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":53,"modele":"cerebras-gpt-13b","nom":"Cerebras-GPT-13B","editeur":"Cerebras Systems","page":"https://quelleia.com/modeles/cerebras-gpt-13b/","score_publie":0.735,"score_0_1":0.47,"niveau_suggere":-67.4,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":54,"modele":"qwen-1-8b","nom":"Qwen-1_8B","editeur":"Alibaba","page":"https://quelleia.com/modeles/qwen-1-8b/","score_publie":0.733,"score_0_1":0.466,"niveau_suggere":-69.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":55,"modele":"gpt-neo-2-7b","nom":"GPT-Neo-2.7B","editeur":"EleutherAI","page":"https://quelleia.com/modeles/gpt-neo-2-7b/","score_publie":0.729,"score_0_1":0.458,"niveau_suggere":-73,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":56,"modele":"gpt-2-1-5b","nom":"GPT-2 (1.5B)","editeur":"OpenAI","page":"https://quelleia.com/modeles/gpt-2-1-5b/","score_publie":0.705,"score_0_1":0.41,"niveau_suggere":-95.6,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":57,"modele":"chatglm2-6b","nom":"chatglm2-6b","editeur":"Z.ai (Zhipu AI)","page":"https://quelleia.com/modeles/chatglm2-6b/","score_publie":0.696,"score_0_1":0.392,"niveau_suggere":-104.2,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":58,"modele":"opt-1-3b","nom":"OPT-1.3B","editeur":"Meta AI","page":"https://quelleia.com/modeles/opt-1-3b/","score_publie":0.69,"score_0_1":0.38,"niveau_suggere":-110.1,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"},{"rang":59,"modele":"stablelm-tuned-alpha-7b","nom":"stablelm-tuned-alpha-7b","editeur":"Stability AI","page":"https://quelleia.com/modeles/stablelm-tuned-alpha-7b/","score_publie":0.658,"score_0_1":0.316,"niveau_suggere":-142.8,"reflexion":"inconnue","harnais":null,"configurations_mesurees":1,"date":null,"source":"https://epoch.ai/benchmarks"}]}