{"source":"Quelle IA, le classement des modèles de langage","page":"https://quelleia.com/modeles/gpt-5-4/","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-5.4"},{"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-5-4","nom":"GPT-5.4","editeur":"OpenAI","pays":"États-Unis","date_sortie":"2026-03-05","nouveau":false,"poids_ouverts":false,"acces":"API access","licence":"proprietary","local":null,"prix_eur_par_million":{"entree":2.1677,"sortie":13.0061},"hebergeable_ue":true,"classe_energie":"E","contexte_jetons":1050000,"entrees":["text","image","pdf"],"niveaux_reflexion":["none","low","medium","high","xhigh"],"connaissances_jusqu_a":"2025-08-31","general":{"score":88.9,"bas":87.9,"haut":89.9,"mesures":50,"poids_couvert":100,"sources":36,"provisoire":false,"historique":false,"rang":20,"groupe":5},"taches":{"raisonnement":{"score":89.4,"bas":87.6,"haut":91.3,"mesures":8,"benchmarks_tache":9,"couverture":89,"provisoire":false,"rang":15},"code":{"score":88.9,"bas":87.5,"haut":90.4,"mesures":9,"benchmarks_tache":13,"couverture":69,"provisoire":false,"rang":23},"agents":{"score":88.4,"bas":84.6,"haut":92.2,"mesures":6,"benchmarks_tache":9,"couverture":67,"provisoire":false,"rang":9},"usage":{"score":88.1,"bas":80,"haut":96.3,"mesures":1,"benchmarks_tache":1,"couverture":100,"provisoire":false,"rang":22},"maths":{"score":89.4,"bas":88.2,"haut":90.7,"mesures":7,"benchmarks_tache":8,"couverture":88,"provisoire":false,"rang":17},"sciences":{"score":89.4,"bas":85,"haut":93.9,"mesures":4,"benchmarks_tache":5,"couverture":80,"provisoire":false,"rang":14},"francais":{"score":83.7,"bas":73.9,"haut":93.5,"mesures":10,"benchmarks_tache":10,"couverture":100,"provisoire":false,"rang":21},"vision":{"score":89.9,"bas":87.1,"haut":92.8,"mesures":3,"benchmarks_tache":3,"couverture":100,"provisoire":false,"rang":11},"ecriture":{"score":87.6,"bas":79.4,"haut":95.8,"mesures":1,"benchmarks_tache":1,"couverture":100,"provisoire":false,"rang":27},"web":{"score":82.4,"bas":77.5,"haut":87.3,"mesures":1,"benchmarks_tache":1,"couverture":100,"provisoire":false,"rang":50}},"reflexions_mesurees":["elevee","faible","maximale","moyenne","sans"],"harnais_mesures":["Codex CLI","ForgeCode","OpenHands","mini-SWE-agent"],"mesures":[{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.9777777777777776,"score_0_1":0.9778,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":88.2},{"benchmark":"euroeval_allocine","tache":"francais","etat":"sature","score_publie":96,"score_0_1":0.96,"reflexion":"sans","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":69.2},{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.9555555555555556,"score_0_1":0.9555,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":83.3},{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.953,"score_0_1":0.953,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-03-06","source":"https://epoch.ai/benchmarks","niveau_suggere":82.9},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.9367,"score_0_1":0.9367,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":89.8},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.9367,"score_0_1":0.9367,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://arcprize.org/leaderboard","niveau_suggere":89.8},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.9267,"score_0_1":0.9267,"reflexion":"elevee","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":88.9},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.933,"score_0_1":0.9107,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-03-06","source":"https://epoch.ai/benchmarks","niveau_suggere":88.1},{"benchmark":"dtbench","tache":"raisonnement","etat":"sature","score_publie":0.944,"score_0_1":0.9067,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://conceptualreasoning.ai/dtbench","niveau_suggere":94.4},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.898989898989899,"score_0_1":0.8653,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":82.6},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.8617,"score_0_1":0.8617,"reflexion":"moyenne","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":84.8},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.8888888888888888,"score_0_1":0.8519,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":81.3},{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.8444444444444444,"score_0_1":0.8443,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":73.9},{"benchmark":"frontiermath","tache":"maths","etat":"actif","score_publie":0.476,"score_0_1":0.8351,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-03-06","source":"https://epoch.ai/benchmarks","niveau_suggere":90},{"benchmark":"terminal_bench","tache":"code","etat":"actif","score_publie":0.818,"score_0_1":0.818,"reflexion":"inconnue","harnais":"ForgeCode","provenance":"externe","date":"2026-03-12","source":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","niveau_suggere":90.2},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.8484848484848485,"score_0_1":0.798,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":76.9},{"benchmark":"frontiermath_tiers_1_3_v2","tache":"maths","etat":"actif","score_publie":0.7859649122807018,"score_0_1":0.786,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-06-11","source":"https://epoch.ai/benchmarks","niveau_suggere":91.7},{"benchmark":"euroeval_hellaswag_fr","tache":"francais","etat":"actif","score_publie":78,"score_0_1":0.78,"reflexion":"sans","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":79.4},{"benchmark":"weirdml","tache":"code","etat":"actif","score_publie":0.777,"score_0_1":0.777,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://htihle.github.io/weirdml.html","niveau_suggere":95.7},{"benchmark":"swe_bench_verified","tache":"code","etat":"actif","score_publie":0.768595041322314,"score_0_1":0.7686,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-03-06","source":"https://epoch.ai/benchmarks","niveau_suggere":84.7},{"benchmark":"euroeval_eltec","tache":"francais","etat":"actif","score_publie":74.59,"score_0_1":0.7459,"reflexion":"sans","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":97.3},{"benchmark":"metr_time_horizons","tache":"agents","etat":"actif","score_publie":0.743443,"score_0_1":0.7434,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":85.8},{"benchmark":"metr_time_horizons","tache":"agents","etat":"actif","score_publie":0.743443,"score_0_1":0.7434,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/","niveau_suggere":85.8},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.7395,"score_0_1":0.7395,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":89.6},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.7395,"score_0_1":0.7395,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":89.6},{"benchmark":"euroeval_include_fr","tache":"francais","etat":"actif","score_publie":72.23,"score_0_1":0.7223,"reflexion":"sans","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":75.5},{"benchmark":"euroeval_fquad","tache":"francais","etat":"actif","score_publie":71.08,"score_0_1":0.7108,"reflexion":"sans","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":76.5},{"benchmark":"arc_agi","tache":"raisonnement","etat":"sature","score_publie":0.6817000000000001,"score_0_1":0.6817,"reflexion":"faible","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":78.5},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.675,"score_0_1":0.675,"reflexion":"elevee","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":88.3},{"benchmark":"gpqa_diamond","tache":"sciences","etat":"actif","score_publie":0.7474747474747475,"score_0_1":0.6633,"reflexion":"sans","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":68.8},{"benchmark":"euroeval_scala_fr","tache":"francais","etat":"actif","score_publie":65,"score_0_1":0.65,"reflexion":"sans","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":99.6},{"benchmark":"arena_expert","tache":"sciences","etat":"actif","score_publie":1515.3,"score_0_1":0.6305,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":88.2},{"benchmark":"arena_maths","tache":"maths","etat":"actif","score_publie":1491.4,"score_0_1":0.6236,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":88.2},{"benchmark":"lmca","tache":"raisonnement","etat":"actif","score_publie":51.97,"score_0_1":0.6114,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://conceptualreasoning.ai/lmca","niveau_suggere":89.6},{"benchmark":"comparia","tache":"francais","etat":"actif","score_publie":1099,"score_0_1":0.6049,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arene.comparia.beta.gouv.fr/ranking","niveau_suggere":105.8},{"benchmark":"arena_francais","tache":"francais","etat":"actif","score_publie":1499.6,"score_0_1":0.6044,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":86.7},{"benchmark":"arena_code","tache":"code","etat":"actif","score_publie":1520.5,"score_0_1":0.6039,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":86.1},{"benchmark":"arena_texte","tache":"usage","etat":"actif","score_publie":1475.2,"score_0_1":0.6038,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":86.8},{"benchmark":"arena_ecriture","tache":"ecriture","etat":"actif","score_publie":1442.4,"score_0_1":0.6013,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":86.1},{"benchmark":"arena_expert","tache":"sciences","etat":"actif","score_publie":1492,"score_0_1":0.5987,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":83.8},{"benchmark":"arena_code","tache":"code","etat":"actif","score_publie":1511.2,"score_0_1":0.5911,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":84},{"benchmark":"arena_ecriture","tache":"ecriture","etat":"actif","score_publie":1434.8,"score_0_1":0.5908,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":84.4},{"benchmark":"arena_texte","tache":"usage","etat":"actif","score_publie":1464.9,"score_0_1":0.5895,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":84.4},{"benchmark":"arena_francais","tache":"francais","etat":"actif","score_publie":1485.6,"score_0_1":0.5849,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":83.7},{"benchmark":"arena_vision","tache":"vision","etat":"actif","score_publie":1284.1,"score_0_1":0.5828,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-28","source":"https://arena.ai/leaderboard","niveau_suggere":89.1},{"benchmark":"arena_maths","tache":"maths","etat":"actif","score_publie":1459.9,"score_0_1":0.5802,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":81.7},{"benchmark":"otis_mock_aime_2024_2025","tache":"maths","etat":"sature","score_publie":0.5777777777777777,"score_0_1":0.5774,"reflexion":"sans","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":64.5},{"benchmark":"arena_vision","tache":"vision","etat":"actif","score_publie":1278.8,"score_0_1":0.5754,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-28","source":"https://arena.ai/leaderboard","niveau_suggere":87.8},{"benchmark":"weirdml","tache":"code","etat":"actif","score_publie":0.5744,"score_0_1":0.5744,"reflexion":"sans","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":79.6},{"benchmark":"scicode","tache":"code","etat":"actif","score_publie":0.565972222222222,"score_0_1":0.566,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://artificialanalysis.ai/evaluations/scicode","niveau_suggere":91},{"benchmark":"proofbench","tache":"maths","etat":"sature","score_publie":0.56,"score_0_1":0.56,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://www.vals.ai/benchmarks/proof_bench","niveau_suggere":88.8},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.5542,"score_0_1":0.5542,"reflexion":"moyenne","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":86},{"benchmark":"arena_webdev","tache":"web","etat":"actif","score_publie":1465.3,"score_0_1":0.5368,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":81.7},{"benchmark":"apex_agents","tache":"agents","etat":"actif","score_publie":0.524,"score_0_1":0.524,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":89.1},{"benchmark":"deepswe","tache":"code","etat":"actif","score_publie":0.5176991150442478,"score_0_1":0.5177,"reflexion":"maximale","harnais":"mini-SWE-agent","provenance":"externe","date":null,"source":"https://deepswe.datacurve.ai/","niveau_suggere":87},{"benchmark":"vending_bench_2","tache":"agents","etat":"actif","score_publie":6144.176,"score_0_1":0.5109,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":88.7},{"benchmark":"arena_webdev","tache":"web","etat":"actif","score_publie":1443.4,"score_0_1":0.5053,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":80.2},{"benchmark":"frontiermath_tier_4_v2","tache":"maths","etat":"sature","score_publie":0.49,"score_0_1":0.49,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-06-11","source":"https://epoch.ai/benchmarks","niveau_suggere":90.1},{"benchmark":"frontiermath_tier_4","tache":"maths","etat":"actif","score_publie":0.271,"score_0_1":0.4517,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-03-06","source":"https://epoch.ai/benchmarks","niveau_suggere":89.1},{"benchmark":"gbaeval","tache":"code","etat":"actif","score_publie":0.4514817826662626,"score_0_1":0.4515,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://gbaeval.com/model/gpt-5-4","niveau_suggere":89.2},{"benchmark":"simpleqa_verified","tache":"donnees","etat":"actif","score_publie":0.451,"score_0_1":0.451,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-08-27","source":"https://epoch.ai/benchmarks","niveau_suggere":84.3},{"benchmark":"arena_webdev","tache":"web","etat":"actif","score_publie":1394.9,"score_0_1":0.4359,"reflexion":"inconnue","harnais":null,"provenance":"independant","date":"2026-09-30","source":"https://arena.ai/leaderboard","niveau_suggere":76.9},{"benchmark":"euroeval_multiloko_fr","tache":"francais","etat":"actif","score_publie":42.88,"score_0_1":0.4288,"reflexion":"sans","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":74.4},{"benchmark":"chess_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.44,"score_0_1":0.4108,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-03-11","source":"https://epoch.ai/benchmarks","niveau_suggere":90.8},{"benchmark":"euroeval_orange_sum","tache":"francais","etat":"actif","score_publie":37.11,"score_0_1":0.3711,"reflexion":"sans","harnais":null,"provenance":"independant","date":null,"source":"https://euroeval.com/leaderboards/Monolingual/french/","niveau_suggere":53.6},{"benchmark":"deepresearch_bench","tache":"agents","etat":"actif","score_publie":0.351,"score_0_1":0.351,"reflexion":"faible","harnais":null,"provenance":"externe","date":null,"source":"https://evals.futuresearch.ai/#drb","niveau_suggere":38.1},{"benchmark":"chess_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.38,"score_0_1":0.3476,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":88.1},{"benchmark":"chess_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.38,"score_0_1":0.3476,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":88.1},{"benchmark":"hle","tache":"sciences","etat":"actif","score_publie":0.3624,"score_0_1":0.3303,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":87.7},{"benchmark":"gso_bench","tache":"code","etat":"actif","score_publie":0.3137,"score_0_1":0.3137,"reflexion":"maximale","harnais":"OpenHands","provenance":"externe","date":null,"source":"https://gso-bench.github.io/index.html","niveau_suggere":87.3},{"benchmark":"mystery_game_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.37,"score_0_1":0.306,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-07-24","source":"https://epoch.ai/benchmarks","niveau_suggere":89.7},{"benchmark":"arc_agi_2","tache":"raisonnement","etat":"sature","score_publie":0.2917,"score_0_1":0.2917,"reflexion":"faible","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":81.1},{"benchmark":"cl_bench","tache":"donnees","etat":"actif","score_publie":0.279,"score_0_1":0.279,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":95.8},{"benchmark":"blueprint_bench_2","tache":"vision","etat":"actif","score_publie":0.2711496746203904,"score_0_1":0.2711,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://andonlabs.com/evals/blueprint-bench-2","niveau_suggere":90.4},{"benchmark":"gso_bench","tache":"code","etat":"actif","score_publie":0.2549,"score_0_1":0.2549,"reflexion":"elevee","harnais":"OpenHands","provenance":"externe","date":null,"source":"https://gso-bench.github.io/index.html","niveau_suggere":85.2},{"benchmark":"ebr_bench","tache":"agents","etat":"actif","score_publie":0.253968,"score_0_1":0.254,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-06-25","source":"https://epoch.ai/benchmarks","niveau_suggere":89.6},{"benchmark":"critpt","tache":"sciences","etat":"actif","score_publie":0.2342857143,"score_0_1":0.2343,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":92.3},{"benchmark":"cl_bench_life","tache":"donnees","etat":"actif","score_publie":0.217,"score_0_1":0.217,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":91.2},{"benchmark":"mystery_game_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.28,"score_0_1":0.2069,"reflexion":"moyenne","harnais":null,"provenance":"independant","date":"2026-08-28","source":"https://epoch.ai/benchmarks","niveau_suggere":86.4},{"benchmark":"cl_bench_life","tache":"donnees","etat":"actif","score_publie":0.193,"score_0_1":0.193,"reflexion":"elevee","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":88.7},{"benchmark":"posttrainbench","tache":"agents","etat":"actif","score_publie":0.19,"score_0_1":0.19,"reflexion":"elevee","harnais":"Codex CLI","provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":79.7},{"benchmark":"forecastbench","tache":"raisonnement","etat":"actif","score_publie":59.5,"score_0_1":0.19,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":71},{"benchmark":"enigmaeval","tache":"raisonnement","etat":"actif","score_publie":0.1596,"score_0_1":0.1596,"reflexion":"maximale","harnais":null,"provenance":"externe","date":null,"source":"https://labs.scale.com/leaderboard/enigma_eval","niveau_suggere":84.2},{"benchmark":"chess_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.2,"score_0_1":0.1582,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":77.5},{"benchmark":"mirrorcode","tache":"code","etat":"actif","score_publie":0.1555555555555555,"score_0_1":0.1556,"reflexion":"elevee","harnais":null,"provenance":"independant","date":"2026-08-10","source":"https://epoch.ai/benchmarks","niveau_suggere":88.5},{"benchmark":"cl_bench_life","tache":"donnees","etat":"actif","score_publie":0.138,"score_0_1":0.138,"reflexion":"inconnue","harnais":null,"provenance":"externe","date":null,"source":"https://epoch.ai/benchmarks","niveau_suggere":81.9},{"benchmark":"furniture_assembly","tache":"vision","etat":"actif","score_publie":0.375,"score_0_1":0.1071,"reflexion":"maximale","harnais":null,"provenance":"independant","date":"2026-09-10","source":"https://epoch.ai/benchmarks","niveau_suggere":90},{"benchmark":"mystery_game_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.17,"score_0_1":0.0857,"reflexion":"faible","harnais":null,"provenance":"independant","date":"2026-08-27","source":"https://epoch.ai/benchmarks","niveau_suggere":79.9},{"benchmark":"mystery_game_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.16,"score_0_1":0.0747,"reflexion":"sans","harnais":null,"provenance":"independant","date":"2026-08-30","source":"https://epoch.ai/benchmarks","niveau_suggere":78.9},{"benchmark":"chess_puzzles","tache":"raisonnement","etat":"actif","score_publie":0.05,"score_0_1":0.0004,"reflexion":"sans","harnais":null,"provenance":"independant","date":"2026-07-15","source":"https://epoch.ai/benchmarks","niveau_suggere":47.6}],"qualite_prix":{"prix_mixte_eur":4.877,"score_attendu":87.1,"avantage":1.8,"avantage_bas":0.8,"avantage_haut":2.8,"meilleur_compromis":false,"rang":47}}}