{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/modeles/gpt-5-2/","edition":"2026-W40","date":"2026-09-28","echelle":"Score IA : 100 pour le meilleur modèle au lancement du site, 50 pour GPT-4o de novembre 2024. bas et haut bornent l'intervalle à 90 %.","licence":"Réutilisation libre en citant « Quelle IA, quelleia.com » et la date de l'édition. Les mesures renvoient à leurs sources d'origine.","modele":{"liens":[{"type":"essayer","libelle":"Essayer sur ChatGPT","url":"https://chatgpt.com"},{"type":"openrouter","libelle":"Utiliser par API sur OpenRouter","url":"https://openrouter.ai/openai/gpt-5.2"},{"type":"api","libelle":"Documentation de l'API (OpenAI)","url":"https://platform.openai.com/docs"},{"type":"editeur","libelle":"Site de l'éditeur (OpenAI)","url":"https://openai.com"}],"id":"gpt-5-2","nom":"GPT-5.2","editeur":"OpenAI","pays":"États-Unis","date_sortie":"2025-12-11","nouveau":false,"poids_ouverts":false,"acces":"API access","licence":"proprietary","local":null,"prix_eur_par_million":{"entree":1.5174,"sortie":12.139},"hebergeable_ue":true,"classe_energie":"E","contexte_jetons":400000,"entrees":["text","image"],"niveaux_reflexion":["none","low","medium","high","xhigh"],"connaissances_jusqu_a":"2025-08-31","general":{"score":84.2,"bas":82.4,"haut":86,"mesures":44,"poids_couvert":100,"sources":30,"provisoire":false,"historique":false,"rang":42,"groupe":6},"taches":{"raisonnement":{"score":85,"bas":81.9,"haut":88,"mesures":9,"benchmarks_tache":9,"couverture":100,"provisoire":false,"rang":33},"code":{"score":84.7,"bas":81.3,"haut":88.2,"mesures":5,"benchmarks_tache":13,"couverture":38,"provisoire":false,"rang":27},"agents":{"score":88.1,"bas":83.4,"haut":92.8,"mesures":5,"benchmarks_tache":9,"couverture":56,"provisoire":false,"rang":10},"usage":{"score":81.6,"bas":73.4,"haut":89.8,"mesures":1,"benchmarks_tache":1,"couverture":100,"provisoire":false,"rang":54},"maths":{"score":84.7,"bas":82.4,"haut":87.1,"mesures":7,"benchmarks_tache":8,"couverture":88,"provisoire":false,"rang":33},"sciences":{"score":83.2,"bas":78.1,"haut":88.2,"mesures":3,"benchmarks_tache":5,"couverture":60,"provisoire":false,"rang":30},"francais":{"score":76.6,"bas":66,"haut":87.3,"mesures":10,"benchmarks_tache":10,"couverture":100,"provisoire":false,"rang":38},"vision":{"score":89.7,"bas":82.2,"haut":97.2,"mesures":2,"benchmarks_tache":3,"couverture":67,"provisoire":false,"rang":14},"ecriture":{"score":79.8,"bas":71.6,"haut":88,"mesures":1,"benchmarks_tache":1,"couverture":100,"provisoire":false,"rang":64},"web":{"score":79,"bas":74.1,"haut":83.8,"mesures":1,"benchmarks_tache":1,"couverture":100,"provisoire":false,"rang":63}},"reflexions_mesurees":["elevee","faible","maximale","moyenne","sans"],"harnais_mesures":["Codex CLI","Droid","Mux","OpenHands","Terminus 2"],"mesures":[{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.9611111111111112,"score_0_1":0.9611,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":84.3},{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.961,"score_0_1":0.961,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2025-12-13","source":"https://epoch.ai/benchmarks","niveau_suggere":84.3},{"benchmark":"euroeval_allocine","tache":"francais","etat":"sature","score_publie":95.69,"score_0_1":0.9569,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":65.5},{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.9388888888888888,"score_0_1":0.9388,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":81.1},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.914,"score_0_1":0.8853,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2025-12-13","source":"https://epoch.ai/benchmarks","niveau_suggere":84.8},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.862,"score_0_1":0.862,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":84.8},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.8617,"score_0_1":0.8617,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://arcprize.org/leaderboard","niveau_suggere":84.8},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.8617,"score_0_1":0.8617,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://arcprize.org/leaderboard","niveau_suggere":84.8},{"benchmark":"dtbench","tache":"raisonnement","etat":"sature","score_publie":0.9093,"score_0_1":0.8488,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://conceptualreasoning.ai/dtbench","niveau_suggere":86.6},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.8819444444444444,"score_0_1":0.8426,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":80.5},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.8787878787878788,"score_0_1":0.8384,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":80.1},{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.7888888888888889,"score_0_1":0.7887,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":71.4},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.7867000000000001,"score_0_1":0.7867,"reflexion":"elevee","harnais":null,"provenance":"externe","date":null,"source":"https://arcprize.org/leaderboard","niveau_suggere":81.7},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.827020202020202,"score_0_1":0.7694,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":75},{"benchmark":"vpct","tache":"vision","etat":"archive","score_publie":0.84,"score_0_1":0.76,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":86.1},{"benchmark":"metr_time_horizons","tache":"agents","etat":"actif","score_publie":0.752875,"score_0_1":0.7529,"reflexion":"elevee","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":86.9},{"benchmark":"swe_bench_verified","tache":"code","etat":"actif","score_publie":0.737603305785124,"score_0_1":0.7376,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-02-12","source":"https://epoch.ai/benchmarks","niveau_suggere":81.5},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.7267,"score_0_1":0.7267,"reflexion":"moyenne","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":79.7},{"benchmark":"weirdml","tache":"code","etat":"actif","score_publie":0.7219,"score_0_1":0.7219,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://htihle.github.io/weirdml.html","niveau_suggere":90.7},{"benchmark":"euroeval_hellaswag_fr","tache":"francais","etat":"actif","score_publie":71.53,"score_0_1":0.7153,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":68.1},{"benchmark":"frontiermath","tache":"maths","etat":"actif","score_publie":0.407,"score_0_1":0.714,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2025-12-13","source":"https://epoch.ai/benchmarks","niveau_suggere":85.7},{"benchmark":"frontiermath","tache":"maths","etat":"actif","score_publie":0.4034482758620689,"score_0_1":0.7078,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":85.6},{"benchmark":"euroeval_fquad","tache":"francais","etat":"actif","score_publie":70.22,"score_0_1":0.7022,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":71.5},{"benchmark":"euroeval_scala_fr","tache":"francais","etat":"actif","score_publie":67.46,"score_0_1":0.6746,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":105.4},{"benchmark":"frontiermath_tiers_1_3_v2","tache":"maths","etat":"actif","score_publie":0.674,"score_0_1":0.674,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-06-11","source":"https://epoch.ai/benchmarks","niveau_suggere":87.3},{"benchmark":"terminal_bench","tache":"code","etat":"actif","score_publie":0.6494382022470001,"score_0_1":0.6494,"reflexion":"inconnue","harnais":"Droid","provenance":"externe","date":"2025-12-11","source":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","niveau_suggere":83.1},{"benchmark":"terminal_bench","tache":"code","etat":"actif","score_publie":0.649,"score_0_1":0.649,"reflexion":"moyenne","harnais":"Droid","provenance":"externe","date":"2025-12-24","source":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","niveau_suggere":83},{"benchmark":"frontiermath","tache":"maths","etat":"actif","score_publie":0.3689655172413793,"score_0_1":0.6473,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":83.9},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.7323232323232324,"score_0_1":0.6431,"reflexion":"sans","harnais":null,"provenance":"independant","date":"2026-07-13","source":"https://epoch.ai/benchmarks","niveau_suggere":67.7},{"benchmark":"euroeval_eltec","tache":"francais","etat":"actif","score_publie":63.94,"score_0_1":0.6394,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":58},{"benchmark":"weirdml","tache":"code","etat":"actif","score_publie":0.6344,"score_0_1":0.6344,"reflexion":"moyenne","harnais":null,"provenance":"externe","date":null,"source":"https://htihle.github.io/weirdml.html","niveau_suggere":83.9},{"benchmark":"terminal_bench","tache":"code","etat":"actif","score_publie":0.629213483146,"score_0_1":0.6292,"reflexion":"inconnue","harnais":"Codex CLI","provenance":"externe","date":"2025-12-11","source":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","niveau_suggere":82.3},{"benchmark":"terminal_bench","tache":"code","etat":"actif","score_publie":0.629,"score_0_1":0.629,"reflexion":"moyenne","harnais":"Codex CLI","provenance":"externe","date":"2025-12-18","source":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","niveau_suggere":82.3},{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.6222222222222222,"score_0_1":0.6218,"reflexion":"sans","harnais":null,"provenance":"independant","date":"2026-07-13","source":"https://epoch.ai/benchmarks","niveau_suggere":65.8},{"benchmark":"terminal_bench","tache":"code","etat":"actif","score_publie":0.606741573034,"score_0_1":0.6067,"reflexion":"inconnue","harnais":"Mux","provenance":"externe","date":"2025-12-11","source":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","niveau_suggere":81.6},{"benchmark":"euroeval_include_fr","tache":"francais","etat":"actif","score_publie":59.85,"score_0_1":0.5985,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":55.4},{"benchmark":"comparia","tache":"francais","etat":"actif","score_publie":1087,"score_0_1":0.5883,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arene.comparia.beta.gouv.fr/ranking","niveau_suggere":100.6},{"benchmark":"arena_francais","tache":"francais","etat":"actif","score_publie":1481.3,"score_0_1":0.5789,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":82.8},{"benchmark":"arena_expert","tache":"sciences","etat":"actif","score_publie":1474.9,"score_0_1":0.5748,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":80.5},{"benchmark":"arena_maths","tache":"maths","etat":"actif","score_publie":1456.1,"score_0_1":0.5748,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":81},{"benchmark":"arena_code","tache":"code","etat":"actif","score_publie":1489,"score_0_1":0.5599,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":79.2},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.5567,"score_0_1":0.5567,"reflexion":"faible","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":75.3},{"benchmark":"arena_francais","tache":"francais","etat":"actif","score_publie":1464.2,"score_0_1":0.5548,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":79.1},{"benchmark":"arena_expert","tache":"sciences","etat":"actif","score_publie":1459.3,"score_0_1":0.5528,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":77.5},{"benchmark":"arena_texte","tache":"usage","etat":"actif","score_publie":1437.3,"score_0_1":0.5506,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":78},{"benchmark":"arena_code","tache":"code","etat":"actif","score_publie":1482,"score_0_1":0.5499,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":77.6},{"benchmark":"arena_texte","tache":"usage","etat":"actif","score_publie":1435.6,"score_0_1":0.5482,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":77.7},{"benchmark":"terminal_bench","tache":"code","etat":"actif","score_publie":0.540449438202,"score_0_1":0.5404,"reflexion":"inconnue","harnais":"Terminus 2","provenance":"externe","date":"2025-12-11","source":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","niveau_suggere":79.4},{"benchmark":"arena_maths","tache":"maths","etat":"actif","score_publie":1431.7,"score_0_1":0.5402,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":76},{"benchmark":"arena_ecriture","tache":"ecriture","etat":"actif","score_publie":1391.5,"score_0_1":0.5295,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":74.7},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.5291,"score_0_1":0.5291,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":85.6},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.5291,"score_0_1":0.5291,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":85.6},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.5291,"score_0_1":0.5291,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":85.6},{"benchmark":"arena_vision","tache":"vision","etat":"actif","score_publie":1242.8,"score_0_1":0.5242,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-28","source":"https://arena.ai/leaderboard","niveau_suggere":79.4},{"benchmark":"arena_ecriture","tache":"ecriture","etat":"actif","score_publie":1387,"score_0_1":0.523,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":73.7},{"benchmark":"lmca","tache":"raisonnement","etat":"actif","score_publie":43.91,"score_0_1":0.5166,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://conceptualreasoning.ai/lmca","niveau_suggere":83.2},{"benchmark":"vpct","tache":"vision","etat":"archive","score_publie":0.67,"score_0_1":0.505,"reflexion":"elevee","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":81.5},{"benchmark":"arena_vision","tache":"vision","etat":"actif","score_publie":1228.7,"score_0_1":0.5039,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-28","source":"https://arena.ai/leaderboard","niveau_suggere":76},{"benchmark":"gdpval","tache":"agents","etat":"archive","score_publie":0.497,"score_0_1":0.497,"reflexion":"sans","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":87.9},{"benchmark":"weirdml","tache":"code","etat":"actif","score_publie":0.496,"score_0_1":0.496,"reflexion":"sans","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":74.3},{"benchmark":"weirdml","tache":"code","etat":"actif","score_publie":0.496,"score_0_1":0.496,"reflexion":"faible","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":74.3},{"benchmark":"vending_bench_2","tache":"agents","etat":"actif","score_publie":3591.326,"score_0_1":0.4924,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":86.8},{"benchmark":"frontiermath","tache":"maths","etat":"actif","score_publie":0.2655172413793103,"score_0_1":0.4658,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":79.4},{"benchmark":"arena_webdev","tache":"web","etat":"actif","score_publie":1415.5,"score_0_1":0.4652,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":78.3},{"benchmark":"chess_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.49,"score_0_1":0.4634,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2025-12-15","source":"https://epoch.ai/benchmarks","niveau_suggere":93},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.4333,"score_0_1":0.4333,"reflexion":"elevee","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":83.8},{"benchmark":"deepresearch_bench","tache":"agents","etat":"actif","score_publie":0.411,"score_0_1":0.411,"reflexion":"faible","harnais":null,"provenance":"externe","date":null,"source":"https://evals.futuresearch.ai/#drb","niveau_suggere":60.7},{"benchmark":"simpleqa_verified","tache":"donnees","etat":"actif","score_publie":0.371,"score_0_1":0.371,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-08-27","source":"https://epoch.ai/benchmarks","niveau_suggere":78.9},{"benchmark":"chess_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.4,"score_0_1":0.3687,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":89},{"benchmark":"chess_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.4,"score_0_1":0.3687,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":89},{"benchmark":"euroeval_orange_sum","tache":"francais","etat":"actif","score_publie":36.77,"score_0_1":0.3677,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":45.3},{"benchmark":"simplebench","tache":"raisonnement","etat":"actif","score_publie":0.458,"score_0_1":0.3496,"reflexion":"elevee","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":73.5},{"benchmark":"simplebench","tache":"raisonnement","etat":"actif","score_publie":0.4579999999999999,"score_0_1":0.3496,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":73.5},{"benchmark":"simpleqa_verified","tache":"donnees","etat":"actif","score_publie":0.343,"score_0_1":0.343,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-08-27","source":"https://epoch.ai/benchmarks","niveau_suggere":77},{"benchmark":"simpleqa_verified","tache":"donnees","etat":"actif","score_publie":0.328,"score_0_1":0.328,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2026-08-27","source":"https://epoch.ai/benchmarks","niveau_suggere":75.9},{"benchmark":"simpleqa_verified","tache":"donnees","etat":"actif","score_publie":0.327,"score_0_1":0.327,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2026-08-27","source":"https://epoch.ai/benchmarks","niveau_suggere":75.8},{"benchmark":"frontiermath_tier_4_v2","tache":"maths","etat":"sature","score_publie":0.317,"score_0_1":0.317,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-06-11","source":"https://epoch.ai/benchmarks","niveau_suggere":87},{"benchmark":"frontiermath_tier_4","tache":"maths","etat":"actif","score_publie":0.188,"score_0_1":0.3133,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":86},{"benchmark":"frontiermath_tier_4","tache":"maths","etat":"actif","score_publie":0.188,"score_0_1":0.3133,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2025-12-14","source":"https://epoch.ai/benchmarks","niveau_suggere":86},{"benchmark":"euroeval_multiloko_fr","tache":"francais","etat":"actif","score_publie":30.49,"score_0_1":0.3049,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":64.9},{"benchmark":"frontiermath_tier_4","tache":"maths","etat":"actif","score_publie":0.167,"score_0_1":0.2783,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":85.1},{"benchmark":"gso_bench","tache":"code","etat":"actif","score_publie":0.2745,"score_0_1":0.2745,"reflexion":"elevee","harnais":"OpenHands","provenance":"externe","date":null,"source":"https://gso-bench.github.io/index.html","niveau_suggere":85.9},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.2667,"score_0_1":0.2667,"reflexion":"moyenne","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":80.6},{"benchmark":"hle","tache":"sciences","etat":"actif","score_publie":0.278,"score_0_1":0.2416,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":82.9},{"benchmark":"ebr_bench","tache":"agents","etat":"actif","score_publie":0.230159,"score_0_1":0.2302,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-06-26","source":"https://epoch.ai/benchmarks","niveau_suggere":88.9},{"benchmark":"forecastbench","tache":"raisonnement","etat":"actif","score_publie":60.1,"score_0_1":0.202,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":77.4},{"benchmark":"chess_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.23,"score_0_1":0.1898,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":79.7},{"benchmark":"cl_bench","tache":"donnees","etat":"actif","score_publie":0.182,"score_0_1":0.182,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":79},{"benchmark":"cl_bench","tache":"donnees","etat":"actif","score_publie":0.181,"score_0_1":0.181,"reflexion":"elevee","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":78.8},{"benchmark":"mystery_game_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.23,"score_0_1":0.1518,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-08-06","source":"https://epoch.ai/benchmarks","niveau_suggere":84},{"benchmark":"proofbench","tache":"maths","etat":"sature","score_publie":0.15,"score_0_1":0.15,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://www.vals.ai/benchmarks/proof_bench","niveau_suggere":81.2},{"benchmark":"mystery_game_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.22,"score_0_1":0.1408,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2026-08-28","source":"https://epoch.ai/benchmarks","niveau_suggere":83.4},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.1233,"score_0_1":0.1233,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://arcprize.org/leaderboard","niveau_suggere":62.4},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.1233,"score_0_1":0.1233,"reflexion":"sans","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":62.4},{"benchmark":"furniture_assembly","tache":"vision","etat":"actif","score_publie":0.3833333333333333,"score_0_1":0.119,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-09-10","source":"https://epoch.ai/benchmarks","niveau_suggere":90.4},{"benchmark":"frontiermath_tier_4","tache":"maths","etat":"actif","score_publie":0.0625,"score_0_1":0.1042,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2025-12-11","source":"https://epoch.ai/benchmarks","niveau_suggere":78.8},{"benchmark":"enigmaeval","tache":"raisonnement","etat":"actif","score_publie":0.1039,"score_0_1":0.1039,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://labs.scale.com/leaderboard/enigma_eval","niveau_suggere":79.2},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.0972,"score_0_1":0.0972,"reflexion":"faible","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":75.2},{"benchmark":"mystery_game_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.14,"score_0_1":0.0527,"reflexion":"sans","harnais":null,"provenance":"independant","date":"2026-08-27","source":"https://epoch.ai/benchmarks","niveau_suggere":76.6},{"benchmark":"remote_labor_index","tache":"agents","etat":"actif","score_publie":0.0250199999999999,"score_0_1":0.025,"reflexion":"moyenne","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":83.1},{"benchmark":"remote_labor_index","tache":"agents","etat":"actif","score_publie":0.0208,"score_0_1":0.0208,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":81.8},{"benchmark":"mystery_game_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.1,"score_0_1":0.0086,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2026-08-27","source":"https://epoch.ai/benchmarks","niveau_suggere":65.7},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.0083,"score_0_1":0.0083,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":64.7},{"benchmark":"chess_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.04,"score_0_1":0,"reflexion":"sans","harnais":null,"provenance":"independant","date":"2026-07-13","source":"https://epoch.ai/benchmarks","niveau_suggere":47.6}],"qualite_prix":{"prix_mixte_eur":4.173,"score_attendu":86.4,"avantage":-2.2,"avantage_bas":-4,"avantage_haut":-0.4,"meilleur_compromis":false,"rang":69}}}