{"id":2102,"date":"2026-08-03T19:59:47","date_gmt":"2026-08-03T19:59:47","guid":{"rendered":"https:\/\/convly.ai\/ai-benchmarks\/"},"modified":"2026-08-03T19:59:47","modified_gmt":"2026-08-03T19:59:47","slug":"ai-benchmarks","status":"publish","type":"page","link":"https:\/\/convly.ai\/ar\/ai-benchmarks\/","title":{"rendered":"\u0634\u0631\u062d \u0645\u0642\u0627\u064a\u064a\u0633 \u0623\u062f\u0627\u0621 \u0627\u0644\u0630\u0643\u0627\u0621 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a (\u0662\u0660\u0662\u0666): \u0645\u0627 \u0627\u0644\u0630\u064a \u064a\u0642\u064a\u0633\u0647 \u0643\u0644 \u0645\u0646\u0647\u0627 \u0648\u0645\u0646 \u064a\u062a\u0635\u062f\u0651\u0631\u0647\u0627\u061f"},"content":{"rendered":"

\u0645\u0642\u0627\u064a\u064a\u0633 \u0623\u062f\u0627\u0621 \u0627\u0644\u0630\u0643\u0627\u0621 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a \u0647\u064a \u0645\u062c\u0645\u0648\u0639\u0629 \u062b\u0627\u0628\u062a\u0629 \u0645\u0646 \u0627\u0644\u0645\u0647\u0627\u0645 \u062a\u064f\u0633\u062a\u062e\u062f\u0645 \u0644\u062a\u0642\u064a\u064a\u0645 \u0642\u062f\u0631\u0629 \u0627\u0644\u0646\u0645\u0648\u0630\u062c\u060c \u0628\u062d\u064a\u062b \u064a\u064f\u0645\u0643\u0646
\n\u0645\u0642\u0627\u0631\u0646\u0629 \u0646\u0645\u0627\u0630\u062c \u0645\u062e\u062a\u0644\u0641\u0629 \u0639\u0644\u0649 \u0646\u0641\u0633 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631.<\/strong> \u0644\u0627 \u064a\u0648\u062c\u062f \u0645\u0642\u064a\u0627\u0633 \u0623\u062f\u0627\u0621 \u0648\u062d\u064a\u062f \u064a\u064f\u0642\u0631\u0651\u0631 \u0623\u064a \u0646\u0645\u0648\u0630\u062c \u0647\u0648 \u0627\u0644\u0623\u0641\u0636\u0644 \u2014 \u0641\u0643\u0644 \u0645\u0642\u064a\u0627\u0633 \u064a\u0642\u064a\u0633 \u062c\u0627\u0646\u0628\u0627\u064b \u0636\u064a\u0651\u0642\u0627\u064b\u060c \u0648\u0628\u0639\u0636 \u0623\u0628\u0631\u0632 \u0627\u0644\u0645\u0642\u0627\u064a\u064a\u0633 \u0627\u0644\u0645\u0639\u0631\u0648\u0641\u0629
\n\u0648\u0635\u0644\u062a \u0627\u0644\u0622\u0646 \u0625\u0644\u0649 \u062f\u0631\u062c\u0629 \u062a\u0634\u0628\u0639\u064d \u0639\u0627\u0644\u064a\u0629. \u0648\u0627\u0644\u0646\u0647\u062c \u0627\u0644\u0639\u0645\u0644\u064a \u0647\u0648 \u0627\u0644\u0627\u0637\u0644\u0627\u0639 \u0639\u0644\u0649 \u0627\u062b\u0646\u064a\u0646 \u0623\u0648 \u062b\u0644\u0627\u062b\u0629 \u0645\u0642\u0627\u064a\u064a\u0633 \u062a\u062a\u0637\u0627\u0628\u0642 \u0645\u0639 \u062d\u0650\u0645\u0644 \u0627\u0644\u0639\u0645\u0644 \u0627\u0644\u0641\u0639\u0644\u064a \u0627\u0644\u062e\u0627\u0635 \u0628\u0643\u060c \u0648\u0645\u0639\u0627\u0645\u0644\u0629 \u0623\u064a \u0631\u0642\u0645 \u0631\u0626\u064a\u0633\u064a \u0648\u0627\u062d\u062f \u0628\u062d\u0630\u0631\u064d \u0634\u062f\u064a\u062f.
\n\u0645\u0646 \u064a\u062a\u0635\u062f\u0651\u0631 \u062d\u0627\u0644\u064a\u0627\u064b\u061f
\n\u0645\u0631\u062a\u0651\u0628 \u0648\u0641\u0642 \u0645\u0624\u0634\u0631 \u0627\u0644\u062a\u062d\u0644\u064a\u0644 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a \u0644\u0644\u0630\u0643\u0627\u0621\u060c \u0648\u0647\u0648 \u0645\u0624\u0634\u0631 \u0645\u0631\u0643\u0651\u0628 \u064a\u062c\u0645\u0639 \u0639\u062f\u0629 \u062a\u0642\u064a\u064a\u0645\u0627\u062a \u0645\u0633\u062a\u0642\u0644\u0629 \u0628\u062f\u0644\u0627\u064b \u0645\u0646 \u0627\u062e\u062a\u0628\u0627\u0631 \u0648\u0627\u062d\u062f. \u0623\u0645\u0627 \u0627\u0644\u0639\u0645\u0648\u062f \u0627\u0644\u0623\u062e\u064a\u0631 \u0641\u0647\u0648 \u0645\u0627 \u062a\u062a\u062c\u0627\u0647\u0644\u0647 \u0645\u0639\u0638\u0645 \u0627\u0644\u0645\u0642\u0627\u0631\u0646\u0627\u062a: \u0627\u0644\u0646\u062a\u064a\u062c\u0629 \u0645\u0642\u0633\u0648\u0645\u0629\u064b \u0639\u0644\u0649 \u0627\u0644\u0633\u0639\u0631 \u0627\u0644\u0645\u062f\u0645\u062c\u060c \u0648\u0647\u0648 \u0645\u0627 \u062a\u0634\u062a\u0631\u064a\u0647 \u0641\u0639\u0644\u064a\u0627\u064b.<\/p>\n

\u0646\u062a\u064a\u062c\u0629 \u0627\u0644\u0630\u0643\u0627\u0621<\/h2>\n

\u0627\u0644\u0646\u062a\u064a\u062c\u0629 \u0644\u0643\u0644 \u062f\u0648\u0644\u0627\u0631
\n\u0627\u0644\u0646\u062a\u0627\u0626\u062c \u0645\u064f\u0639\u0637\u0627\u0629 \u0648\u0641\u0642 \u0645\u0642\u064a\u0627\u0633 \u0645\u0624\u0634\u0631 \u0627\u0644\u062a\u062d\u0644\u064a\u0644 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a \u0644\u0644\u0630\u0643\u0627\u0621\u060c \u0648\u0647\u0648 \u0645\u0624\u0634\u0631 \u0645\u0631\u0643\u0651\u0628 \u064a\u062c\u0645\u0639 \u0639\u062f\u0629 \u062a\u0642\u064a\u064a\u0645\u0627\u062a \u0645\u0633\u062a\u0642\u0644\u0629 \u0628\u062f\u0644\u0627\u064b \u0645\u0646 \u0623\u064a \u0627\u062e\u062a\u0628\u0627\u0631 \u0648\u062d\u064a\u062f \u2014 \u0648\u0647\u0630\u0627 \u0647\u0648 \u0627\u0644\u0646\u0647\u062c \u0627\u0644\u0635\u062d\u064a\u062d \u0644\u062a\u0641\u0633\u064a\u0631 \u0623\u064a \u0627\u062f\u0639\u0627\u0621\u064d \u0625\u0639\u0644\u0627\u0645\u064a \u0639\u0646 \u00ab\u0623\u0641\u0636\u0644 \u0646\u0645\u0648\u0630\u062c\u00bb. \u0648\u0623\u0642\u0648\u0649 \u0646\u0645\u0648\u0630\u062c \u0630\u064a \u0623\u0648\u0632\u0627\u0646 \u0645\u0641\u062a\u0648\u062d\u0629 \u0627\u0644\u0645\u0635\u062f\u0631 \u0647\u0648
\n\u0628\u0646\u062a\u064a\u062c\u0629 \u0665\u0667. \u0648\u064a\u0645\u0643\u0646\u0643 \u062a\u0631\u062a\u064a\u0628 \u0627\u0644\u0645\u062c\u0645\u0648\u0639\u0629 \u0627\u0644\u0643\u0627\u0645\u0644\u0629 \u0628\u0646\u0641\u0633\u0643 \u0639\u0628\u0631<\/p>\n

\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n
#<\/th>\u0627\u0644\u0646\u0645\u0648\u0630\u062c<\/th>\u0627\u0644\u0645\u0637\u0648\u0651\u0631<\/th>\u0645\u0642\u0627\u064a\u064a\u0633 \u0627\u0644\u0623\u062f\u0627\u0621 \u0627\u0644\u062a\u064a \u062a\u0647\u0645\u0651 \u0641\u064a \u0639\u0627\u0645 \u0662\u0660\u0662\u0666<\/th>\n \u0627\u0644\u0645\u062a\u0648\u0633\u0637 \u0627\u0644\u0645\u064f\u062f\u0645\u062c \u0628\u0627\u0644\u062f\u0648\u0644\u0627\u0631 \u0644\u0643\u0644 \u0645\u0644\u064a\u0648\u0646 \u062f\u0648\u0644\u0627\u0631<\/th>\u0625\u0645 \u0625\u0645 \u0625\u0644 \u064a\u0648 (\u0627\u0644\u0641\u0647\u0645 \u0627\u0644\u0644\u063a\u0648\u064a \u0645\u062a\u0639\u062f\u062f \u0627\u0644\u0645\u0647\u0627\u0645 \u0627\u0644\u0636\u062e\u0645)<\/th><\/tr><\/thead>\n
1<\/td>\n \u0643\u0644\u0627\u0648\u062f \u0623\u0648\u0628\u0633 \u0665<\/a><\/td>\n \u0623\u0646\u062b\u0631\u0648\u0628\u064a\u0643<\/td>\n 61<\/strong><\/td>\n $9.00<\/td>\n 6.8<\/td>\n <\/tr>\n
2<\/td>\n \u0643\u0644\u0627\u0648\u062f \u0641\u0627\u0628\u0644 \u0665<\/a><\/td>\n \u0623\u0646\u062b\u0631\u0648\u0628\u064a\u0643<\/td>\n 60<\/strong><\/td>\n $18.00<\/td>\n 3.3<\/td>\n <\/tr>\n
3<\/td>\n \u062c\u064a \u0628\u064a \u062a\u064a-\u0665.\u0666 \u0633\u0648\u0644<\/a><\/td>\n \u0623\u0648\u0628\u0646 \u0625\u064a\u0647 \u0622\u064a<\/td>\n 59<\/strong><\/td>\n $10.00<\/td>\n 5.9<\/td>\n <\/tr>\n
4<\/td>\n \u0643\u064a\u0645\u064a \u0643\u0663<\/a> \u0623\u0648\u0632\u0627\u0646 \u0645\u0641\u062a\u0648\u062d\u0629 \u0627\u0644\u0645\u0635\u062f\u0631<\/span><\/td>\n \u0645\u0648\u0646 \u0634\u0648\u062a \u0622\u064a<\/td>\n 57<\/strong><\/td>\n $5.40<\/td>\n 10.6<\/td>\n <\/tr>\n
5<\/td>\n \u0643\u0644\u0627\u0648\u062f \u0623\u0648\u0628\u0633 \u0664.\u0668<\/a><\/td>\n \u0623\u0646\u062b\u0631\u0648\u0628\u064a\u0643<\/td>\n 55.7<\/strong><\/td>\n $9.00<\/td>\n 6.2<\/td>\n <\/tr>\n
6<\/td>\n \u062c\u064a \u0628\u064a \u062a\u064a-\u0665.\u0665<\/a><\/td>\n \u0623\u0648\u0628\u0646 \u0625\u064a\u0647 \u0622\u064a<\/td>\n 54.8<\/strong><\/td>\n $10.00<\/td>\n 5.5<\/td>\n <\/tr>\n
7<\/td>\n \u063a\u0644\u0652\u0645 \u0665.\u0662<\/a> \u0623\u0648\u0632\u0627\u0646 \u0645\u0641\u062a\u0648\u062d\u0629 \u0627\u0644\u0645\u0635\u062f\u0631<\/span><\/td>\n \u0632\u0647\u064a \u0628\u0648 \u0622\u064a<\/td>\n 51.1<\/strong><\/td>\n $2.00<\/td>\n 25.6<\/td>\n <\/tr>\n
8<\/td>\n \u062c\u0645\u064a\u0646\u064a \u0663.\u0665 \u0641\u0644\u0627\u0634<\/a><\/td>\n \u062c\u0648\u062c\u0644<\/td>\n 50.2<\/strong><\/td>\n $3.00<\/td>\n 16.7<\/td>\n <\/tr>\n
9<\/td>\n \u0643\u0644\u0627\u0648\u062f \u0633\u0648\u0646\u064a\u062a \u0664.\u0666<\/a><\/td>\n \u0623\u0646\u062b\u0631\u0648\u0628\u064a\u0643<\/td>\n 47<\/strong><\/td>\n $5.40<\/td>\n 8.7<\/td>\n <\/tr>\n
10<\/td>\n \u062c\u0645\u064a\u0646\u064a \u0663.\u0661 \u0628\u0631\u0648<\/a><\/td>\n \u062c\u0648\u062c\u0644<\/td>\n 46.5<\/strong><\/td>\n $4.00<\/td>\n 11.6<\/td>\n <\/tr>\n
11<\/td>\n \u062f\u064a\u0628 \u0633\u064a\u0643 \u0641\u064a4-\u0628\u0631\u0648<\/a> \u0623\u0648\u0632\u0627\u0646 \u0645\u0641\u062a\u0648\u062d\u0629 \u0627\u0644\u0645\u0635\u062f\u0631<\/span><\/td>\n DeepSeek<\/td>\n 44.3<\/strong><\/td>\n $0.522<\/td>\n 84.9<\/td>\n <\/tr>\n
12<\/td>\n \u0643\u064a\u0645\u064a \u0643\u0662.\u0667 \u0643\u0648\u062f<\/a> \u0623\u0648\u0632\u0627\u0646 \u0645\u0641\u062a\u0648\u062d\u0629 \u0627\u0644\u0645\u0635\u062f\u0631<\/span><\/td>\n \u0645\u0648\u0646 \u0634\u0648\u062a \u0622\u064a<\/td>\n 42<\/strong><\/td>\n $0.980<\/td>\n 42.9<\/td>\n <\/tr>\n <\/tbody>\n <\/table>\n

\u0623\u0633\u0626\u0644\u0629 \u0627\u062e\u062a\u064a\u0627\u0631 \u0645\u0646 \u0645\u062a\u0639\u062f\u062f \u062a\u063a\u0637\u064a \u0665\u0667 \u0645\u0648\u0636\u0648\u0639\u0627\u064b\u060c \u0645\u0646 \u0627\u0644\u0631\u064a\u0627\u0636\u064a\u0627\u062a \u0627\u0644\u0627\u0628\u062a\u062f\u0627\u0626\u064a\u0629 \u0625\u0644\u0649 \u0627\u0644\u0642\u0627\u0646\u0648\u0646 \u0627\u0644\u0645\u0647\u0646\u064a. \u0648\u0642\u062f \u0643\u0627\u0646 \u0644\u0633\u0646\u0648\u0627\u062a \u0639\u062f\u064a\u062f\u0629 \u0627\u0644\u0631\u0642\u0645 \u0627\u0644\u0631\u0626\u064a\u0633\u064a \u0627\u0644\u0642\u064a\u0627\u0633\u064a\u060c \u0648\u0647\u0630\u0647 \u0647\u064a \u0627\u0644\u0645\u0634\u0643\u0644\u0629 \u0627\u0644\u064a\u0648\u0645: \u0625\u0630 \u062a\u062a\u0643\u062a\u0651\u0644 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0645\u062a\u0642\u062f\u0645\u0629 \u0641\u064a \u0627\u0644\u0642\u0645\u0629 \u0628\u0634\u0643\u0644\u064d \u0634\u062f\u064a\u062f \u0627\u0644\u062a\u0642\u0627\u0631\u0628 \u0644\u062f\u0631\u062c\u0629 \u0623\u0646 \u0627\u0644\u0641\u0631\u0648\u0642 \u0628\u064a\u0646\u0647\u0627 \u062a\u0642\u0639 \u062f\u0627\u062e\u0644 \u0647\u0627\u0645\u0634 \u0627\u0644\u062e\u0637\u0623. \u0648\u0644\u0627 \u064a\u0632\u0627\u0644 \u0645\u0642\u064a\u0627\u0633 \u0625\u0645 \u0625\u0645 \u0625\u0644 \u064a\u0648 \u0645\u0641\u064a\u062f\u0627\u064b \u0644\u0645\u0642\u0627\u0631\u0646\u0629 \u0646\u0645\u0648\u0630\u062c \u0635\u063a\u064a\u0631 \u0630\u064a \u0623\u0648\u0632\u0627\u0646 \u0645\u0641\u062a\u0648\u062d\u0629 \u0645\u0639 \u0646\u0645\u0648\u0630\u062c \u0643\u0628\u064a\u0631\u060c \u0644\u0643\u0646\u0647 \u064a\u0643\u0627\u062f \u064a\u0643\u0648\u0646 \u0639\u062f\u064a\u0645 \u0627\u0644\u0641\u0627\u0626\u062f\u0629 \u0641\u064a \u0627\u0644\u062a\u0645\u064a\u064a\u0632 \u0628\u064a\u0646 \u0646\u0645\u0648\u0630\u062c\u064a\u0646 \u0645\u062a\u0642\u062f\u0645\u064a\u0646. \u0644\u0630\u0627 \u0627\u0639\u062a\u0628\u0631 \u0623\u064a \u0627\u062f\u0639\u0627\u0621\u064d \u0639\u0646 \u0646\u062a\u064a\u062c\u0629 \u0625\u0645 \u0625\u0645 \u0625\u0644 \u064a\u0648 \u0644\u0639\u0627\u0645 \u0662\u0660\u0662\u0666 \u0628\u0645\u062b\u0627\u0628\u0629 \u0627\u062e\u062a\u0628\u0627\u0631 \u062d\u062f \u0623\u062f\u0646\u0649\u060c \u0648\u0644\u064a\u0633 \u0645\u0639\u064a\u0627\u0631\u0627\u064b \u0644\u0644\u062a\u0631\u062a\u064a\u0628.\n \u0643\u064a\u0645\u064a \u0643\u0663<\/a>\n \u062c \u0628\u064a \u0643\u064a\u0648 \u0625\u064a\u0647 (\u0623\u0633\u0626\u0644\u0629 \u0648\u0625\u062c\u0627\u0628\u0627\u062a \u0639\u0644\u0649 \u0645\u0633\u062a\u0648\u0649 \u0627\u0644\u062f\u0631\u0627\u0633\u0627\u062a \u0627\u0644\u0639\u0644\u064a\u0627 \u0644\u0627 \u064a\u0645\u0643\u0646 \u0627\u0644\u0625\u062c\u0627\u0628\u0629 \u0639\u0646\u0647\u0627 \u0639\u0628\u0631 \u062c\u0648\u062c\u0644)\n \u0644\u0648\u062d\u0629 \u062a\u0635\u0646\u064a\u0641 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0644\u063a\u0648\u064a\u0629 \u0627\u0644\u0643\u0628\u064a\u0631\u0629 (LLM)<\/a>.<\/p>\n<\/div>\n\n

\u0643\u062a\u0628\u0647\u0627 \u062f\u0643\u0627\u062a\u0631\u0629 \u0645\u062a\u062e\u0635\u0635\u0648\u0646 \u0641\u064a \u0645\u062c\u0627\u0644\u0627\u062a \u0627\u0644\u0623\u062d\u064a\u0627\u0621 \u0648\u0627\u0644\u0641\u064a\u0632\u064a\u0627\u0621 \u0648\u0627\u0644\u0643\u064a\u0645\u064a\u0627\u0621\u060c \u0648\u0635\u064f\u0645\u0651\u0645\u062a \u0639\u0645\u062f\u0627\u064b \u0628\u062d\u064a\u062b \u0644\u0627 \u064a\u0633\u0627\u0639\u062f \u0627\u0644\u0628\u062d\u062b \u0639\u0628\u0631 \u0627\u0644\u0648\u064a\u0628 \u0641\u064a \u0627\u0644\u0625\u062c\u0627\u0628\u0629 \u0639\u0646\u0647\u0627. \u0648\u0647\u064a \u062a\u062c\u064a\u0628 \u0639\u0646 \u0633\u0624\u0627\u0644 \u0645\u062e\u062a\u0644\u0641 \u0639\u0646 \u0633\u0624\u0627\u0644 \u0625\u0645 \u0625\u0645 \u0625\u0644 \u064a\u0648: \u0641\u0647\u064a \u0644\u0627 \u062a\u0633\u0623\u0644 \u00ab\u0647\u0644 \u0642\u0631\u0623 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0643\u062b\u064a\u0631\u0627\u064b\u061f\u00bb \u0628\u0644 \u062a\u0633\u0623\u0644 \u00ab\u0647\u0644 \u064a\u0633\u062a\u0637\u064a\u0639 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0627\u0644\u0627\u0633\u062a\u0646\u062a\u0627\u062c \u0648\u0627\u0644\u062a\u0641\u0643\u064a\u0631 \u0641\u064a \u0645\u0634\u0643\u0644\u0629 \u0635\u0639\u0628\u0629 \u062d\u0642\u0627\u064b \u0636\u0645\u0646 \u0645\u062c\u0627\u0644 \u062a\u0642\u0646\u064a\u061f\u00bb. \u0648\u0628\u0645\u0627 \u0623\u0646\u0647\u0627 \u062a\u0642\u0627\u0648\u0645 \u0643\u0644\u0627\u064b \u0645\u0646 \u0627\u0644\u062d\u0641\u0638 \u0648\u0627\u0644\u0628\u062d\u062b \u0648\u0627\u0644\u0627\u0633\u062a\u0631\u062c\u0627\u0639\u060c \u0641\u0642\u062f \u062d\u0627\u0641\u0638\u062a \u0639\u0644\u0649 \u0642\u064a\u0645\u062a\u0647\u0627 \u0643\u0623\u062f\u0627\u0629 \u062a\u0645\u064a\u064a\u0632 \u0623\u0641\u0636\u0644 \u0645\u0646 \u0625\u0645 \u0625\u0645 \u0625\u0644 \u064a\u0648 \u0645\u0639 \u0645\u0631\u0648\u0631 \u0627\u0644\u0632\u0645\u0646.<\/h2>\n

\u062a\u0642\u0627\u0631\u064a\u0631 \u0623\u062e\u0637\u0627\u0621 \u0641\u0639\u0644\u064a\u0629 \u0645\u0646 \u0645\u0633\u062a\u0648\u062f\u0639\u0627\u062a \u062c\u064a\u062a \u0647\u0627\u0628 \u0627\u0644\u062d\u0642\u064a\u0642\u064a\u0629\u060c \u0648\u062a\u064f\u0642\u064a\u064e\u0651\u0645 \u0628\u0646\u0627\u0621\u064b \u0639\u0644\u0649 \u0645\u0627 \u0625\u0630\u0627 \u0643\u0627\u0646\u062a \u0627\u0644\u062a\u0635\u062d\u064a\u062d\u0627\u062a \u0627\u0644\u062a\u064a \u0642\u062f\u0651\u0645\u0647\u0627 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u062a\u062c\u0639\u0644 \u0627\u0644\u0628\u0631\u0646\u0627\u0645\u062c<\/h3>\n

\u0623\u0633\u0626\u0644\u0629 \u0645\u062a\u0639\u062f\u062f\u0629 \u0627\u0644\u062e\u064a\u0627\u0631\u0627\u062a \u0639\u0628\u0631 57 \u0645\u0648\u0636\u0648\u0639\u064b\u0627\u060c \u062a\u0628\u062f\u0623 \u0645\u0646 \u0627\u0644\u0631\u064a\u0627\u0636\u064a\u0627\u062a \u0627\u0644\u0627\u0628\u062a\u062f\u0627\u0626\u064a\u0629 \u0648\u062a\u0646\u062a\u0647\u064a \u0628\u0627\u0644\u0645\u0648\u0627\u0636\u064a\u0639 \u0627\u0644\u0645\u0647\u0646\u064a\u0629
\n\u0648\u0627\u0644\u0642\u0627\u0646\u0648\u0646. \u0648\u0644\u0633\u0646\u0648\u0627\u062a\u064d \u0639\u062f\u064a\u062f\u0629 \u0643\u0627\u0646\u062a \u0627\u0644\u0646\u062a\u064a\u062c\u0629 \u0627\u0644\u0627\u0641\u062a\u0631\u0627\u0636\u064a\u0629 \u0627\u0644\u0631\u0626\u064a\u0633\u064a\u0629 \u0627\u0644\u0645\u064f\u0639\u0644\u064e\u0651\u0642\u0629 \u0641\u064a \u0647\u0630\u0627 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u060c \u0648\u0627\u0644\u0622\u0646 \u062a\u0643\u0645\u0646 \u0627\u0644\u0645\u0634\u0643\u0644\u0629 \u0641\u064a \u0623\u0646 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0631\u0627\u0626\u062f\u0629 \u062a\u062a\u062c\u0645\u0639 \u0628\u0625\u062d\u0643\u0627\u0645 \u0634\u062f\u064a\u062f \u0639\u0646\u062f \u0627\u0644\u0642\u0645\u0629 \u0628\u062d\u064a\u062b \u062a\u0642\u0639 \u0627\u0644\u0641\u0631\u0648\u0642 \u0628\u064a\u0646\u0647\u0627 \u062f\u0627\u062e\u0644 \u0646\u0637\u0627\u0642 \u0627\u0644\u0636\u0648\u0636\u0627\u0621 \u0627\u0644\u0625\u062d\u0635\u0627\u0626\u064a\u0629. \u0648\u0644\u0627 \u064a\u0632\u0627\u0644 \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0645\u0641\u064a\u062f\u064b\u0627 \u0644\u0645\u0642\u0627\u0631\u0646\u0629 \u0646\u0645\u0648\u0630\u062c \u0635\u063a\u064a\u0631 \u0645\u0641\u062a\u0648\u062d \u0627\u0644\u0645\u0635\u062f\u0631 \u0645\u0639 \u0646\u0645\u0648\u0630\u062c \u0643\u0628\u064a\u0631 \u062c\u062f\u064b\u0651\u0627\u060c \u0644\u0643\u0646\u0647 \u064a\u0643\u0627\u062f \u064a\u0643\u0648\u0646 \u0639\u062f\u064a\u0645 \u0627\u0644\u0641\u0627\u0626\u062f\u0629 \u0641\u064a \u0627\u0644\u062a\u0645\u064a\u064a\u0632 \u0628\u064a\u0646 \u0646\u0645\u0648\u0630\u062c\u064a\u0646 \u0631\u0627\u0626\u062f\u064a\u0646. \u0648\u0639\u0644\u064a\u0647\u060c \u0641\u0639\u0644\u064a\u0643 \u0627\u0639\u062a\u0628\u0627\u0631 \u0623\u064a \u0627\u062f\u0639\u0627\u0621\u064d \u0645\u062a\u0639\u0644\u0642\u064d \u0628\u0646\u0633\u0628\u0629 \u0627\u0644\u0623\u062f\u0627\u0621 \u0641\u064a \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0644\u0639\u0627\u0645 2026 \u0628\u0645\u062b\u0627\u0628\u0629 \u0627\u062e\u062a\u0628\u0627\u0631\u064d \u0644\u0644\u062d\u062f \u0627\u0644\u0623\u062f\u0646\u0649 \u0645\u0646 \u0627\u0644\u0623\u062f\u0627\u0621 \u0644\u0627 \u0643\u0623\u062f\u0627\u0629\u064d \u0644\u062a\u0631\u062a\u064a\u0628 \u0627\u0644\u0646\u0645\u0627\u0630\u062c.
\n\u064a\u0638\u0644 \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0645\u0641\u064a\u062f\u064b\u0627 \u0644\u0648\u0636\u0639 \u0646\u0645\u0648\u0630\u062c \u0635\u063a\u064a\u0631 \u0645\u0641\u062a\u0648\u062d \u0627\u0644\u0645\u0635\u062f\u0631 \u0645\u0642\u0627\u0628\u0644 \u0646\u0645\u0648\u0630\u062c \u0643\u0628\u064a\u0631 \u062c\u062f\u064b\u0651\u0627\u060c \u0644\u0643\u0646\u0647 \u064a\u0643\u0627\u062f \u064a\u0643\u0648\u0646 \u0639\u062f\u064a\u0645 \u0627\u0644\u0641\u0627\u0626\u062f\u0629 \u0641\u064a \u0627\u0644\u062a\u0645\u064a\u064a\u0632 \u0628\u064a\u0646 \u0646\u0645\u0648\u0630\u062c\u064a\u0646 \u0631\u0627\u0626\u062f\u064a\u0646. \u0648\u0639\u0644\u064a\u0647\u060c \u0641\u0639\u0644\u064a\u0643 \u0627\u0639\u062a\u0628\u0627\u0631 \u0623\u064a \u0627\u062f\u0639\u0627\u0621\u064d \u0645\u062a\u0639\u0644\u0642\u064d \u0628\u0646\u0633\u0628\u0629 \u0627\u0644\u0623\u062f\u0627\u0621 \u0641\u064a \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0644\u0639\u0627\u0645 2026 \u0628\u0645\u062b\u0627\u0628\u0629 \u0627\u062e\u062a\u0628\u0627\u0631\u064d \u0644\u0644\u062d\u062f \u0627\u0644\u0623\u062f\u0646\u0649 \u0645\u0646 \u0627\u0644\u0623\u062f\u0627\u0621 \u0644\u0627 \u0643\u0623\u062f\u0627\u0629\u064d \u0644\u062a\u0631\u062a\u064a\u0628 \u0627\u0644\u0646\u0645\u0627\u0630\u062c.
\n\u064a\u0638\u0644 \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0645\u0641\u064a\u062f\u064b\u0627 \u0644\u0648\u0636\u0639 \u0646\u0645\u0648\u0630\u062c \u0635\u063a\u064a\u0631 \u0645\u0641\u062a\u0648\u062d \u0627\u0644\u0645\u0635\u062f\u0631 \u0645\u0642\u0627\u0628\u0644 \u0646\u0645\u0648\u0630\u062c \u0643\u0628\u064a\u0631 \u062c\u062f\u064b\u0651\u0627\u060c \u0644\u0643\u0646\u0647 \u064a\u0643\u0627\u062f \u064a\u0643\u0648\u0646 \u0639\u062f\u064a\u0645 \u0627\u0644\u0641\u0627\u0626\u062f\u0629 \u0641\u064a \u0627\u0644\u062a\u0645\u064a\u064a\u0632 \u0628\u064a\u0646 \u0646\u0645\u0648\u0630\u062c\u064a\u0646 \u0631\u0627\u0626\u062f\u064a\u0646. \u0648\u0639\u0644\u064a\u0647\u060c \u0641\u0639\u0644\u064a\u0643 \u0627\u0639\u062a\u0628\u0627\u0631 \u0623\u064a \u0627\u062f\u0639\u0627\u0621\u064d \u0645\u062a\u0639\u0644\u0642\u064d \u0628\u0646\u0633\u0628\u0629 \u0627\u0644\u0623\u062f\u0627\u0621 \u0641\u064a \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0644\u0639\u0627\u0645 2026 \u0628\u0645\u062b\u0627\u0628\u0629 \u0627\u062e\u062a\u0628\u0627\u0631\u064d \u0644\u0644\u062d\u062f \u0627\u0644\u0623\u062f\u0646\u0649 \u0645\u0646 \u0627\u0644\u0623\u062f\u0627\u0621 \u0644\u0627 \u0643\u0623\u062f\u0627\u0629\u064d \u0644\u062a\u0631\u062a\u064a\u0628 \u0627\u0644\u0646\u0645\u0627\u0630\u062c.
\n\u064a\u0638\u0644 \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0645\u0641\u064a\u062f\u064b\u0627 \u0644\u0648\u0636\u0639 \u0646\u0645\u0648\u0630\u062c \u0635\u063a\u064a\u0631 \u0645\u0641\u062a\u0648\u062d \u0627\u0644\u0645\u0635\u062f\u0631 \u0645\u0642\u0627\u0628\u0644 \u0646\u0645\u0648\u0630\u062c \u0643\u0628\u064a\u0631 \u062c\u062f\u064b\u0651\u0627\u060c \u0644\u0643\u0646\u0647 \u064a\u0643\u0627\u062f \u064a\u0643\u0648\u0646 \u0639\u062f\u064a\u0645 \u0627\u0644\u0641\u0627\u0626\u062f\u0629 \u0641\u064a \u0627\u0644\u062a\u0645\u064a\u064a\u0632 \u0628\u064a\u0646 \u0646\u0645\u0648\u0630\u062c\u064a\u0646 \u0631\u0627\u0626\u062f\u064a\u0646. \u0648\u0639\u0644\u064a\u0647\u060c \u0641\u0639\u0644\u064a\u0643 \u0627\u0639\u062a\u0628\u0627\u0631 \u0623\u064a \u0627\u062f\u0639\u0627\u0621\u064d \u0645\u062a\u0639\u0644\u0642\u064d \u0628\u0646\u0633\u0628\u0629 \u0627\u0644\u0623\u062f\u0627\u0621 \u0641\u064a \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0644\u0639\u0627\u0645 2026 \u0628\u0645\u062b\u0627\u0628\u0629 \u0627\u062e\u062a\u0628\u0627\u0631\u064d \u0644\u0644\u062d\u062f \u0627\u0644\u0623\u062f\u0646\u0649 \u0645\u0646 \u0627\u0644\u0623\u062f\u0627\u0621 \u0644\u0627 \u0643\u0623\u062f\u0627\u0629\u064d \u0644\u062a\u0631\u062a\u064a\u0628 \u0627\u0644\u0646\u0645\u0627\u0630\u062c.<\/p>\n

GPQA (\u0623\u0633\u0626\u0644\u0629 \u0648\u0623\u062c\u0648\u0628\u0629 \u0639\u0644\u0649 \u0645\u0633\u062a\u0648\u0649 \u0627\u0644\u062f\u0631\u0627\u0633\u0627\u062a \u0627\u0644\u0639\u0644\u064a\u0627\u060c \u063a\u064a\u0631 \u0642\u0627\u0628\u0644\u0629 \u0644\u0644\u0628\u062d\u062b \u0639\u0628\u0631 \u0645\u062d\u0631\u0643 \u062c\u0648\u062c\u0644)<\/h3>\n

\u0643\u062a\u0628\u0647\u0627 \u062e\u0628\u0631\u0627\u0621 \u062d\u0627\u0635\u0644\u0648\u0646 \u0639\u0644\u0649 \u062f\u0631\u062c\u0629 \u0627\u0644\u062f\u0643\u062a\u0648\u0631\u0627\u0647 \u0641\u064a \u0645\u062c\u0627\u0644\u0627\u062a \u0627\u0644\u0623\u062d\u064a\u0627\u0621 \u0648\u0627\u0644\u0641\u064a\u0632\u064a\u0627\u0621 \u0648\u0627\u0644\u0643\u064a\u0645\u064a\u0627\u0621\u060c \u0648\u0635\u064f\u0645\u0650\u0651\u0645\u062a \u0639\u0645\u062f\u0627\u064b \u0628\u062d\u064a\u062b \u0644\u0627 \u064a\u0633\u0627\u0639\u062f \u0627\u0644\u0628\u062d\u062b \u0639\u0628\u0631 \u0627\u0644\u0648\u064a\u0628 \u0641\u064a \u0625\u064a\u062c\u0627\u062f \u0627\u0644\u0625\u062c\u0627\u0628\u0627\u062a.
\n\u0648\u064a\u062c\u064a\u0628 \u0647\u0630\u0627 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631 \u0639\u0646 \u0633\u0624\u0627\u0644\u064d \u0645\u062e\u062a\u0644\u0641\u064d \u0639\u0646 \u0627\u0644\u0633\u0624\u0627\u0644 \u0627\u0644\u0630\u064a \u064a\u0637\u0631\u062d\u0647 \u0627\u062e\u062a\u0628\u0627\u0631 MMLU: \u0641\u0647\u0648 \u0644\u0627 \u064a\u0633\u0623\u0644 \u00ab\u0647\u0644 \u0642\u0631\u0623 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0643\u062b\u064a\u0631\u064b\u0627\u061f\u00bb \u0628\u0644 \u064a\u0633\u0623\u0644 \u00ab\u0647\u0644 \u064a\u0633\u062a\u0637\u064a\u0639 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0627\u0644\u0627\u0633\u062a\u0646\u062a\u0627\u062c \u0627\u0644\u0645\u0646\u0637\u0642\u064a \u0644\u062d\u0644 \u0645\u0634\u0643\u0644\u0629\u064d \u0628\u0627\u0644\u063a\u0629 \u0627\u0644\u0635\u0639\u0648\u0628\u0629 \u0641\u064a \u0645\u062c\u0627\u0644\u064d \u062a\u0642\u0646\u064a\u064d\u0651 \u0645\u0639\u064a\u064e\u0651\u0646\u061f\u00bb.
\n\u0648\u064a\u062c\u064a\u0628 \u0647\u0630\u0627 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631 \u0639\u0646 \u0633\u0624\u0627\u0644\u064d \u0645\u062e\u062a\u0644\u0641\u064d \u0639\u0646 \u0627\u0644\u0633\u0624\u0627\u0644 \u0627\u0644\u0630\u064a \u064a\u0637\u0631\u062d\u0647 \u0627\u062e\u062a\u0628\u0627\u0631 MMLU: \u0641\u0647\u0648 \u0644\u0627 \u064a\u0633\u0623\u0644 \u00ab\u0647\u0644 \u0642\u0631\u0623 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0643\u062b\u064a\u0631\u064b\u0627\u061f\u00bb \u0628\u0644 \u064a\u0633\u0623\u0644 \u00ab\u0647\u0644 \u064a\u0633\u062a\u0637\u064a\u0639 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0627\u0644\u0627\u0633\u062a\u0646\u062a\u0627\u062c \u0627\u0644\u0645\u0646\u0637\u0642\u064a \u0644\u062d\u0644 \u0645\u0634\u0643\u0644\u0629\u064d \u0628\u0627\u0644\u063a\u0629 \u0627\u0644\u0635\u0639\u0648\u0628\u0629 \u0641\u064a \u0645\u062c\u0627\u0644\u064d \u062a\u0642\u0646\u064a\u064d\u0651 \u0645\u0639\u064a\u064e\u0651\u0646\u061f\u00bb.
\n\u0648\u0628\u0645\u0627 \u0623\u0646 \u0647\u0630\u0627 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631 \u064a\u0642\u0627\u0648\u0645 \u0643\u0644\u064b\u0651\u0627 \u0645\u0646 \u0627\u0644\u062d\u0641\u0638 \u0627\u0644\u0622\u0644\u064a \u0648\u0627\u0633\u062a\u0631\u062c\u0627\u0639 \u0627\u0644\u0645\u0639\u0644\u0648\u0645\u0627\u062a\u060c \u0641\u0642\u062f \u0638\u0644 \u0623\u0643\u062b\u0631 \u062b\u0628\u0627\u062a\u064b\u0627 \u0645\u0646 \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0643\u0623\u062f\u0627\u0629\u064d \u062a\u0645\u064a\u064a\u0632\u064a\u0629\u064d \u0645\u0639 \u0645\u0631\u0648\u0631 \u0627\u0644\u0632\u0645\u0646.<\/p>\n

\u0633\u0648-\u0628\u0646\u0634 \u0645\u064f\u062d\u0642\u064e\u0651\u0642<\/h3>\n

\u062a\u0642\u0627\u0631\u064a\u0631 \u0623\u062e\u0637\u0627\u0621 \u0641\u0639\u0644\u064a\u0629 \u0645\u0633\u062a\u062e\u0644\u0635\u0629 \u0645\u0646 \u0645\u0633\u062a\u0648\u062f\u0639\u0627\u062a GitHub \u0627\u0644\u062d\u0642\u064a\u0642\u064a\u0629\u060c \u0648\u064a\u062a\u0645 \u062a\u0642\u064a\u064a\u0645\u0647\u0627 \u0628\u0646\u0627\u0621\u064b \u0639\u0644\u0649 \u0645\u0627 \u0625\u0630\u0627 \u0643\u0627\u0646\u062a \u0627\u0644\u062a\u0639\u062f\u064a\u0644\u0627\u062a \u0627\u0644\u062a\u064a \u0642\u062f\u064e\u0651\u0645\u0647\u0627 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u062a\u064f\u0635\u0644\u062d \u0627\u0644\u062e\u0637\u0623 \u0623\u0645 \u0644\u0627.
\n\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0627\u0644\u0645\u0634\u0631\u0648\u0639 \u0627\u0644\u062e\u0627\u0635\u0629 \u062a\u0645\u0631 \u0628\u0646\u062c\u0627\u062d. \u0623\u0645\u0627 \u0645\u062c\u0645\u0648\u0639\u0629 \u00ab\u0627\u0644\u0645\u064f\u062d\u0642\u064e\u0651\u0642\u00bb \u0641\u0647\u064a \u0627\u0644\u062c\u0632\u0621 \u0627\u0644\u0630\u064a \u062a\u0645 \u0627\u0644\u062a\u062d\u0642\u0642 \u0645\u0646 \u0635\u062d\u062a\u0647 \u064a\u062f\u0648\u064a\u064b\u0651\u0627\u060c \u0628\u0639\u062f \u062a\u0631\u0634\u064a\u062d\u0647 \u0644\u0625\u0632\u0627\u0644\u0629
\n\u0627\u0644\u0645\u0647\u0627\u0645 \u0627\u0644\u0645\u0639\u0637\u0648\u0628\u0629 \u0623\u0648 \u0627\u0644\u0645\u0633\u062a\u062d\u064a\u0644\u0629 \u2014 \u0648\u0644\u0630\u0644\u0643 \u064a\u062c\u0628 \u0623\u0646 \u062a\u062a\u062d\u0642\u0642 \u0645\u0646 \u0630\u0644\u0643 \u0623\u064a\u0651<\/em> SWE-bench
\n\u064a\u0642\u062f\u0650\u0651\u0645\u0647 \u0627\u0644\u0645\u0648\u0631\u0650\u0651\u062f. \u0648\u0647\u0630\u0647 \u0647\u064a \u0623\u0643\u062b\u0631 \u0627\u0644\u0645\u0642\u0627\u064a\u064a\u0633 \u0635\u0644\u0629\u064b \u0628\u0627\u0644\u0642\u0631\u0627\u0631 \u0639\u0646\u062f \u0627\u062e\u062a\u064a\u0627\u0631\u0643 \u0646\u0645\u0648\u0630\u062c\u064b\u0627 \u0644\u0648\u0643\u064a\u0644 \u0628\u0631\u0645\u062c\u064a (coding agent)\u060c \u0644\u0623\u0646 \u0627\u0644\u0645\u0647\u0645\u0629 \u0648\u0646\u0638\u0627\u0645 \u0627\u0644\u062a\u0642\u064a\u064a\u0645 \u0643\u0644\u064a\u0647\u0645\u0627 \u0623\u0635\u064a\u0644\u0627\u0646. \u0648\u064a\u0628\u0644\u063a \u0623\u062f\u0627\u0621 \u0646\u0645\u0648\u0630\u062c Claude Sonnet 5
\n85.2% \u0639\u0644\u0649 \u0645\u062c\u0645\u0648\u0639\u0629 SWE-bench \u0627\u0644\u0645\u064f\u062d\u0642\u064e\u0651\u0642\u0629\u060c \u064863.2% \u0639\u0644\u0649 \u0645\u062c\u0645\u0648\u0639\u0629 SWE-bench Pro \u0627\u0644\u0623\u0635\u0639\u0628.
\nTerminal-Bench<\/p>\n

\u0645\u0647\u0627\u0645 \u0648\u0643\u064a\u0644\u064a\u0629 \u062a\u064f\u0646\u0641\u064e\u0651\u0630 \u0641\u064a \u0648\u0627\u062c\u0647\u0629 \u0637\u0631\u0641\u064a\u0629 \u062d\u0642\u064a\u0642\u064a\u0629: \u0645\u062b\u0644 \u062a\u062b\u0628\u064a\u062a \u0628\u0631\u0646\u0627\u0645\u062c \u0645\u0627\u060c \u0623\u0648 \u062a\u0634\u062e\u064a\u0635 \u0639\u0637\u0644\u064d \u0645\u0627\u060c \u0623\u0648 \u0643\u062a\u0627\u0628\u0629 \u0646\u0635 \u0628\u0631\u0645\u062c\u064a \u0644\u0625\u0635\u0644\u0627\u062d \u0645\u0634\u0643\u0644\u0629 \u0645\u0627.<\/h3>\n

\u0648\u0647\u064a \u062a\u0642\u0627\u0633 \u0645\u0647\u0627\u0631\u0629 \u0645\u062e\u062a\u0644\u0641\u0629 \u0639\u0646 \u0643\u062a\u0627\u0628\u0629 \u062a\u0635\u062d\u064a\u062d \u0628\u0631\u0645\u062c\u064a (patch) \u2014 \u0648\u0647\u064a \u0627\u0644\u0642\u062f\u0631\u0629 \u0639\u0644\u0649 \u0627\u0644\u062a\u062d\u0645\u064f\u0651\u0644 \u0623\u062b\u0646\u0627\u0621 \u0625\u0646\u062c\u0627\u0632 \u0645\u0647\u0627\u0645 \u0645\u062a\u0639\u062f\u062f\u0629 \u0627\u0644\u062e\u0637\u0648\u0627\u062a\u060c \u062d\u064a\u062b \u064a\u062c\u0628 \u0639\u0644\u0649 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0642\u0631\u0627\u0621\u0629 \u0645\u062e\u0631\u062c\u0627\u062a \u0627\u0644\u0623\u062e\u0637\u0627\u0621 \u0627\u0644\u062a\u064a \u064a\u0648\u0644\u062f\u0647\u0627 \u0628\u0646\u0641\u0633\u0647 \u0648\u0627\u0644\u062a\u0639\u0627\u0641\u064a \u0645\u0646\u0647\u0627.
\n\u0648\u0623\u0628\u0644\u063a \u0646\u0645\u0648\u0630\u062c Claude Sonnet 5 \u0639\u0646 \u0646\u062a\u064a\u062c\u0629 80.4% \u0639\u0644\u0649 Terminal-Bench 2.1. \u0641\u0625\u0630\u0627 \u0643\u0627\u0646 \u0627\u0633\u062a\u062e\u062f\u0627\u0645\u0643 \u0627\u0644\u0645\u0642\u0635\u0648\u062f \u0647\u0648 \u0648\u0643\u064a\u0644 \u0645\u0633\u062a\u0642\u0644 (autonomous agent) \u0628\u062f\u0644\u064b\u0627 \u0645\u0646 \u0645\u0633\u0627\u0639\u062f \u0628\u0631\u0645\u062c\u064a \u062a\u0641\u0627\u0639\u0644\u064a (inline code assistant)\u060c \u0641\u0625\u0646 \u0647\u0630\u0647 \u0627\u0644\u0645\u062c\u0645\u0648\u0639\u0629 \u0648OSWorld \u062a\u0648\u0641\u0631\u0627\u0646 \u0644\u0643 \u0645\u0639\u0644\u0648\u0645\u0627\u062a\u064d \u0623\u0643\u062b\u0631 \u0641\u0627\u0626\u062f\u0629\u064b \u0645\u0646 \u062a\u0644\u0643 \u0627\u0644\u062a\u064a \u062a\u0642\u062f\u0650\u0651\u0645\u0647\u0627 SWE-bench.
\n\u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0627\u0644\u062d\u0627\u0633\u0648\u0628 \u0641\u064a \u0628\u064a\u0626\u0629 \u0633\u0637\u062d \u0645\u0643\u062a\u0628 \u062d\u0642\u064a\u0642\u064a\u0629 \u2014 \u0645\u062b\u0644 \u0641\u062a\u062d \u0627\u0644\u062a\u0637\u0628\u064a\u0642\u0627\u062a\u060c \u0648\u0627\u0644\u0646\u0642\u0631 \u0639\u0628\u0631 \u0648\u0627\u062c\u0647\u0627\u062a \u0627\u0644\u0645\u0633\u062a\u062e\u062f\u0645\u060c \u0648\u0625\u062a\u0645\u0627\u0645 \u0645\u0647\u0645\u0629 \u0642\u062f \u064a\u0642\u0648\u0645 \u0628\u0647\u0627 \u0627\u0644\u0625\u0646\u0633\u0627\u0646 \u0628\u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0627\u0644\u0641\u0623\u0631\u0629.
\n\u0648\u0627\u0644\u0646\u062a\u0627\u0626\u062c \u0647\u0646\u0627 \u0645\u0646\u062e\u0641\u0636\u0629\u064c \u062c\u062f\u064b\u0651\u0627 \u0645\u0642\u0627\u0631\u0646\u0629\u064b \u0628\u0627\u0644\u0645\u0642\u0627\u064a\u064a\u0633 \u0627\u0644\u0646\u0635\u064a\u0629 \u0639\u0644\u0649 \u0627\u0645\u062a\u062f\u0627\u062f \u0627\u0644\u0645\u062c\u0627\u0644 \u0628\u0623\u0633\u0631\u0647\u060c \u0648\u0647\u064a \u0625\u0634\u0627\u0631\u0629 \u0635\u0627\u062f\u0642\u0629\u064c \u0625\u0644\u0649 \u0645\u062f\u0649 \u062d\u062f\u0627\u062b\u0629 (\u0623\u0648 \u0628\u062f\u0627\u0626\u064a\u0629) \u0648\u0643\u0644\u0627\u0621 \u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0627\u0644\u062d\u0627\u0633\u0648\u0628 \u062d\u062a\u0649 \u0627\u0644\u0622\u0646.
\n\u0648\u0623\u0628\u0644\u063a \u0646\u0645\u0648\u0630\u062c Claude Sonnet 5 \u0639\u0646 \u0646\u062a\u064a\u062c\u0629 81.2% \u0639\u0644\u0649 OSWorld-Verified\u061b \u0628\u064a\u0646\u0645\u0627 \u0623\u0628\u0644\u063a \u0646\u0645\u0648\u0630\u062c Nemotron 3 Nano Omni \u0645\u0646 \u0634\u0631\u0643\u0629 NVIDIA\u060c \u0648\u0647\u0648 \u0646\u0645\u0648\u0630\u062c \u0645\u0641\u062a\u0648\u062d \u0627\u0644\u062d\u062c\u0645 30 \u0645\u0644\u064a\u0627\u0631 \u0645\u0639\u0644\u064e\u0651\u0645\u0629\u060c \u0639\u0646 \u0646\u062a\u064a\u062c\u0629 47.4% \u0639\u0644\u0649 OSWorld\u060c \u0648\u0647\u064a \u0646\u062a\u064a\u062c\u0629 \u0645\u0639\u0642\u0648\u0644\u0629\u064c \u0628\u0627\u0644\u0646\u0633\u0628\u0629 \u0644\u062d\u062c\u0645\u0647.<\/p>\n

\u0623\u0648 \u0625\u0633 \u0648\u0648\u0631\u0644\u062f<\/h3>\n

ARC-AGI
\n\u0623\u0644\u063a\u0627\u0632 \u0627\u0633\u062a\u062f\u0644\u0627\u0644 \u0628\u0635\u0631\u064a \u0645\u062c\u0631\u064e\u0651\u062f \u0645\u064f\u0635\u0645\u064e\u0651\u0645\u0629 \u0628\u062d\u064a\u062b \u0644\u0627 \u064a\u064f\u0645\u0643\u0646 \u0644\u0644\u0646\u0645\u0630\u062c\u0629 \u0627\u0644\u0642\u0627\u0626\u0645\u0629 \u0639\u0644\u0649 \u0645\u0637\u0627\u0628\u0642\u0629 \u0627\u0644\u0623\u0646\u0645\u0627\u0637 \u0641\u064a \u0628\u064a\u0627\u0646\u0627\u062a \u0627\u0644\u062a\u062f\u0631\u064a\u0628 \u0623\u0646 \u062a\u0646\u062a\u0642\u0644 (transfer) \u0625\u0644\u064a\u0647\u0627.
\n\u0648\u062a\u062a\u0643\u0648\u064e\u0651\u0646 \u0643\u0644 \u0645\u0647\u0645\u0629 \u0645\u0646 \u0639\u062f\u062f \u0642\u0644\u064a\u0644 \u0645\u0646 \u062a\u062d\u0648\u064a\u0644\u0627\u062a \u0627\u0644\u0634\u0628\u0643\u0629 (grid transformations) \u0627\u0644\u062a\u064a \u064a\u062c\u0628 \u0639\u0644\u0649 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0627\u0633\u062a\u0646\u062a\u0627\u062c\u0647\u0627 \u0645\u0646 \u0632\u0648\u062c\u064d \u0623\u0648 \u0627\u062b\u0646\u064a\u0646 \u0645\u0646 \u0627\u0644\u0623\u0645\u062b\u0644\u0629.
\n\u0648\u064a\u0639\u062a\u0628\u0631 ARC-AGI \u0623\u0642\u0631\u0628 \u0645\u0627 \u0648\u0635\u0644 \u0625\u0644\u064a\u0647 \u0627\u0644\u0645\u062c\u0627\u0644 \u0625\u0644\u0649 \u0627\u062e\u062a\u0628\u0627\u0631\u064d \u062d\u0642\u064a\u0642\u064a\u064d\u0651 \u0644\u0644\u062a\u0639\u0645\u064a\u0645 \u0627\u0644\u062d\u0642\u064a\u0642\u064a (genuine generalisation) \u0628\u062f\u0644\u064b\u0627 \u0645\u0646 \u0627\u0644\u0627\u0633\u062a\u0631\u062c\u0627\u0639 (recall)\u060c \u0648\u0644\u0630\u0644\u0643 \u0641\u0625\u0646 \u0627\u0644\u062a\u0642\u062f\u064f\u0651\u0645 \u0641\u064a\u0647 \u0628\u0637\u064a\u0621\u064c \u062c\u062f\u064b\u0651\u0627\u060c \u0648\u062a\u064f\u0639\u0627\u0645\u064e\u0644 \u0623\u064a \u0642\u0641\u0632\u0627\u062a \u0646\u0648\u0639\u064a\u0629 \u0641\u064a\u0647 \u0628\u0627\u0639\u062a\u0628\u0627\u0631\u0647\u0627 \u0630\u0627\u062a \u0623\u0647\u0645\u064a\u0629 \u0643\u0628\u064a\u0631\u0629.
\n\u0648\u0623\u064f\u0628\u0644\u063a \u0639\u0646 \u0623\u062f\u0627\u0621 \u0646\u0645\u0648\u0630\u062c Claude Opus 5 \u0639\u0644\u0649 ARC-AGI 3 \u0628\u0623\u0646\u0647 \u064a\u0639\u0627\u062f\u0644 \u062a\u0642\u0631\u064a\u0628\u064b\u0627 \u062b\u0644\u0627\u062b\u0629 \u0623\u0636\u0639\u0627\u0641 \u0623\u0641\u0636\u0644 \u0646\u0645\u0648\u0630\u062c \u062a\u0627\u0644\u064a \u0644\u0647.<\/p>\n

\u0623\u0633\u0626\u0644\u0629 \u0648\u0636\u0639\u0647\u0627 \u062e\u0628\u0631\u0627\u0621 \u0641\u064a \u062a\u062e\u0635\u0635\u0627\u062a \u0639\u062f\u064a\u062f\u0629\u060c \u0648\u0635\u064f\u0645\u0650\u0651\u0645\u062a \u062e\u0635\u064a\u0635\u064b\u0627 \u0644\u0645\u0642\u0627\u0648\u0645\u0629 \u0627\u0644\u062a\u0634\u0628\u0639 (saturation) \u0627\u0644\u0630\u064a \u0623\u0635\u0627\u0628 \u0645\u0642\u064a\u0627\u0633 MMLU.<\/h3>\n

\u0648\u0627\u0644\u0647\u062f\u0641 \u0627\u0644\u062a\u0635\u0645\u064a\u0645\u064a \u0647\u0648 \u0625\u0646\u0634\u0627\u0621 \u0645\u0642\u064a\u0627\u0633 \u064a\u0638\u0644 \u0635\u0639\u0628\u064b\u0627 \u0645\u0639 \u062a\u062d\u0633\u064f\u0651\u0646 \u0627\u0644\u0646\u0645\u0627\u0630\u062c\u060c \u0648\u0628\u0627\u0644\u062a\u0627\u0644\u064a \u062a\u0643\u0648\u0646 \u0627\u0644\u0646\u062a\u0627\u0626\u062c \u0645\u0646\u062e\u0641\u0636\u0629\u064b \u0628\u0637\u0628\u064a\u0639\u062a\u0647\u0627\u060c \u0648\u062a\u0643\u062a\u0633\u0628 \u0627\u0644\u0645\u0643\u0627\u0633\u0628 \u0627\u0644\u0635\u063a\u064a\u0631\u0629 \u062f\u0644\u0627\u0644\u0629\u064b \u0643\u0628\u064a\u0631\u0629\u064b.
\n\u0627\u0642\u0631\u0623 \u0647\u0630\u0627 \u0627\u0644\u0645\u0642\u064a\u0627\u0633 \u0628\u0627\u0639\u062a\u0628\u0627\u0631\u0647 \u0645\u064f\u0645\u064a\u0650\u0651\u0632\u064b\u0627 \u0644\u0644\u062d\u062f\u0648\u062f \u0627\u0644\u0623\u0645\u0627\u0645\u064a\u0629 (frontier discriminator)\u060c \u0648\u0644\u064a\u0633 \u0643\u0645\u0642\u064a\u0627\u0633\u064d \u0644\u0645\u062f\u0649 \u0641\u0627\u0626\u062f\u062a\u0647 \u0641\u064a \u0627\u0644\u0639\u0645\u0644 \u0627\u0644\u0639\u0627\u062f\u064a.
\n\u0644\u064a\u0633 \u0627\u062e\u062a\u0628\u0627\u0631\u064b\u0627 \u0628\u062d\u062f \u0630\u0627\u062a\u0647\u060c \u0628\u0644 \u0647\u0648 \u062a\u0631\u0643\u064a\u0628\u0629 (composite) \u062a\u062f\u0645\u062c \u0639\u062f\u0629 \u062a\u0642\u064a\u064a\u0645\u0627\u062a \u0645\u0633\u062a\u0642\u0644\u0629 \u0641\u064a \u062f\u0631\u062c\u0629 \u0648\u0627\u062d\u062f\u0629.
\n\u0648\u0647\u0630\u0627 \u0647\u0648 \u0642\u064a\u0645\u062a\u0647: \u0641\u0645\u0646 \u0627\u0644\u0633\u0647\u0644 \u0627\u0633\u062a\u0647\u062f\u0627\u0641 \u0623\u064a \u0645\u0642\u064a\u0627\u0633 \u0641\u0631\u062f\u064a\u060c \u0628\u064a\u0646\u0645\u0627 \u064a\u0635\u0639\u0628 \u062c\u062f\u064b\u0651\u0627 \u0627\u0644\u062a\u0644\u0627\u0639\u0628 \u0628\u0645\u0632\u064a\u062c\u064d \u0645\u0646\u0647.
\n\u0648\u0647\u0648 \u0627\u0644\u0645\u0642\u064a\u0627\u0633 \u0627\u0644\u0645\u0633\u062a\u062e\u062f\u0645 \u0641\u064a \u0627\u0644\u062c\u062f\u0648\u0644 \u0623\u0639\u0644\u0627\u0647. \u0648\u064a\u062a\u0635\u062f\u064e\u0651\u0631 \u0646\u0645\u0648\u0630\u062c Claude Opus 5 \u0627\u0644\u062a\u0631\u062a\u064a\u0628 \u062d\u0627\u0644\u064a\u064b\u0651\u0627 \u0628\u0646\u062a\u064a\u062c\u0629 61\u060c \u064a\u0644\u064a\u0647 Claude Fable 5 \u0628\u0646\u062a\u064a\u062c\u0629 60\u060c \u062b\u0645 GPT-5.6 Sol \u0628\u0646\u062a\u064a\u062c\u0629 59 \u2014 \u0628\u064a\u0646\u0645\u0627 \u062d\u0642\u0642 \u0646\u0645\u0648\u0630\u062c Kimi K3 \u0646\u062a\u064a\u062c\u0629 57\u060c \u0648\u0647\u064a \u0623\u0641\u0636\u0644 \u0646\u062a\u064a\u062c\u0629 \u0645\u0633\u062c\u064e\u0651\u0644\u0629 \u062d\u062a\u0649 \u0627\u0644\u0622\u0646 \u0628\u064a\u0646 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0645\u0641\u062a\u0648\u062d\u0629 \u0627\u0644\u0648\u0632\u0646 (open-weight)\u060c \u0648\u062a\u0642\u062a\u0631\u0628 \u0645\u0646 \u0623\u0641\u0636\u0644 \u0646\u0645\u0648\u0630\u062c \u0645\u063a\u0644\u0642 (closed model) \u0628\u0641\u0627\u0631\u0642 \u0623\u0631\u0628\u0639 \u0646\u0642\u0627\u0637 \u0641\u0642\u0637.<\/p>\n

\u0627\u0645\u062a\u062d\u0627\u0646 \u0627\u0644\u0628\u0634\u0631\u064a\u0629 \u0627\u0644\u0623\u062e\u064a\u0631<\/h3>\n

\u0643\u064a\u0641 \u062a\u0642\u0631\u0623 \u0627\u062f\u0639\u0627\u0621\u064b \u062d\u0648\u0644 \u0646\u062a\u064a\u062c\u0629 \u0627\u062e\u062a\u0628\u0627\u0631 \u0623\u062f\u0627\u0621 (benchmark claim)
\n\u0627\u0633\u0623\u0644: \u0645\u0646 \u0627\u0644\u0630\u064a \u0623\u062c\u0631\u0649 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u061f
\n\u0641\u0627\u0644\u0646\u062a\u064a\u062c\u0629 \u0627\u0644\u062a\u064a \u064a\u0639\u0644\u0646 \u0639\u0646\u0647\u0627 \u0627\u0644\u0645\u0648\u0631\u0650\u0651\u062f \u0639\u0644\u0649 \u0635\u0641\u062d\u0629 \u0625\u0637\u0644\u0627\u0642 \u0645\u0646\u062a\u062c\u0647 \u062a\u062e\u062a\u0644\u0641 \u062a\u0645\u0627\u0645\u064b\u0627 \u0639\u0646 \u0627\u0644\u062a\u0642\u064a\u064a\u0645 \u0627\u0644\u0645\u0633\u062a\u0642\u0644\u060c \u0625\u0630 \u0625\u0646 \u0627\u0644\u0644\u0648\u062d\u0627\u062a \u0627\u0644\u0642\u064a\u0627\u062f\u064a\u0629 \u0627\u0644\u0645\u0633\u062a\u0642\u0644\u0629 \u062a\u0637\u0628\u0650\u0651\u0642 \u0646\u0641\u0633 \u0627\u0644\u0625\u0637\u0627\u0631 (harness) \u0639\u0644\u0649 \u062c\u0645\u064a\u0639 \u0627\u0644\u0646\u0645\u0627\u0630\u062c\u060c \u0628\u064a\u0646\u0645\u0627 \u064a\u062e\u062a\u0627\u0631 \u0627\u0644\u0645\u0648\u0631\u0650\u0651\u062f \u0638\u0631\u0648\u0641\u0647 \u0627\u0644\u062e\u0627\u0635\u0629.
\n\u062a\u062d\u0642\u064e\u0651\u0642 \u0645\u0646 \u0648\u062c\u0648\u062f \u062a\u0644\u0648\u062b (contamination).<\/p>\n

\u0645\u0624\u0634\u0631 \u0627\u0644\u062a\u062d\u0644\u064a\u0644 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a \u0644\u0644\u0630\u0643\u0627\u0621<\/h3>\n

\u0625\u0630\u0627 \u0643\u0627\u0646\u062a \u0623\u0633\u0626\u0644\u0629 \u0627\u0644\u0645\u0642\u064a\u0627\u0633 \u0642\u062f \u0633\u0628\u0642\u062a \u062a\u0627\u0631\u064a\u062e \u0625\u063a\u0644\u0627\u0642 \u0628\u064a\u0627\u0646\u0627\u062a \u0627\u0644\u062a\u062f\u0631\u064a\u0628 \u0627\u0644\u062e\u0627\u0635\u0629 \u0628\u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0648\u0648\u062c\u062f\u062a \u0639\u0644\u0646\u064b\u0627 \u0641\u064a \u0645\u0643\u0627\u0646\u064d \u0645\u0627\u060c \u0641\u0645\u0646 \u0627\u0644\u0645\u0631\u062c\u062d \u0623\u0646 \u062a\u0643\u0648\u0646 \u0628\u0639\u0636\u0647\u0627 \u0645\u0648\u062c\u0648\u062f\u0629\u064b \u0641\u064a \u0628\u064a\u0627\u0646\u0627\u062a \u0627\u0644\u062a\u062f\u0631\u064a\u0628.
\n\u0648\u0647\u0630\u0627 \u0647\u0648 \u0627\u0644\u0633\u0628\u0628 \u0627\u0644\u0631\u0626\u064a\u0633\u064a \u0641\u064a \u0623\u0646 \u0627\u0644\u0645\u0642\u0627\u064a\u064a\u0633 \u0627\u0644\u0642\u062f\u064a\u0645\u0629 \u062a\u0645\u064a\u0644 \u0625\u0644\u0649 \u0627\u0644\u062a\u0636\u062e\u064f\u0651\u0645 \u0645\u0639 \u0645\u0631\u0648\u0631 \u0627\u0644\u0648\u0642\u062a\u060c \u0648\u0644\u0630\u0644\u0643 \u0635\u064f\u0645\u0650\u0651\u0645\u062a \u0627\u0644\u0645\u0642\u0627\u064a\u064a\u0633 \u0627\u0644\u062c\u062f\u064a\u062f\u0629 \u0644\u062a\u0643\u0648\u0646 \u00ab\u0645\u062d\u0635\u064e\u0651\u0646\u0629 \u0636\u062f \u062c\u0648\u062c\u0644\u00bb (Google-proof) \u0623\u0648 \u064a\u062a\u0645 \u0627\u0644\u0627\u062d\u062a\u0641\u0627\u0638 \u0628\u0647\u0627 \u0628\u0639\u064a\u062f\u064b\u0627 \u062a\u0645\u0627\u0645\u064b\u0627 \u0639\u0646 \u0628\u064a\u0627\u0646\u0627\u062a \u0627\u0644\u062a\u062f\u0631\u064a\u0628.
\n\u0627\u0646\u062a\u0628\u0647 \u0625\u0644\u0649 \u062a\u062d\u0648\u064f\u0651\u0644 \u062e\u0637 \u0627\u0644\u0646\u0647\u0627\u064a\u0629 (moved goalpost).
\n\u0645\u062b\u0644 \u0630\u0643\u0631 \u00abSWE-bench\u00bb \u062f\u0648\u0646 \u062a\u062d\u062f\u064a\u062f \u00ab\u0627\u0644\u0645\u064f\u062d\u0642\u064e\u0651\u0642\u0629\u00bb\u060c \u0623\u0648 \u0627\u0644\u0625\u0634\u0627\u0631\u0629 \u0625\u0644\u0649 \u0645\u062c\u0645\u0648\u0639\u0629 \u0641\u0631\u0639\u064a\u0629 \u063a\u064a\u0631 \u0645\u0633\u0645\u064e\u0651\u0627\u0629\u060c \u0623\u0648 \u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0639\u062f\u062f \u0645\u062e\u062a\u0644\u0641 \u0645\u0646 \u0627\u0644\u0645\u062d\u0627\u0648\u0644\u0627\u062a\u060c \u0623\u0648 \u0645\u0642\u0627\u0631\u0646\u0629 \u0646\u062a\u064a\u062c\u0629 \u062a\u0634\u0645\u0644 \u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0623\u062f\u0648\u0627\u062a (tool use) \u0645\u0639 \u0623\u062e\u0631\u0649 \u0644\u0627 \u062a\u0634\u0645\u0644\u0647\u0627 \u2014 \u0647\u0630\u0647 \u0647\u064a \u0627\u0644\u0637\u0631\u0642 \u0627\u0644\u0645\u0639\u062a\u0627\u062f\u0629 \u0627\u0644\u062a\u064a \u062a\u062a\u0648\u0642\u0641 \u0641\u064a\u0647\u0627 \u0627\u0644\u0645\u0642\u0627\u0631\u0646\u0629 \u0639\u0646 \u0623\u0646 \u062a\u0643\u0648\u0646 \u00ab\u0645\u0646 \u0646\u0638\u064a\u0631\u064d \u0644\u0646\u0638\u064a\u0631\u00bb (like for like).
\n\u0627\u062d\u0630\u0631 \u0627\u0644\u062a\u0634\u0628\u0639 (saturation).<\/p>\n

\u0641\u0639\u0646\u062f\u0645\u0627 \u064a\u062d\u0635\u0644 \u0643\u0644 \u0646\u0645\u0648\u0630\u062c \u062c\u0627\u062f \u0639\u0644\u0649 \u0646\u062a\u064a\u062c\u0629 \u062a\u0632\u064a\u062f \u0639\u0646 90 \u0641\u064a \u0627\u062e\u062a\u0628\u0627\u0631\u064d \u0645\u0627\u060c \u0641\u0647\u0630\u0627 \u064a\u0639\u0646\u064a \u0623\u0646 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631 \u062a\u0648\u0642\u064e\u0651\u0641 \u0639\u0646 \u0627\u0644\u062a\u0645\u064a\u064a\u0632 \u0628\u064a\u0646 \u0627\u0644\u0646\u0645\u0627\u0630\u062c.<\/h2>\n

\u0648\u0628\u0627\u0644\u062a\u0627\u0644\u064a \u0641\u0625\u0646 \u0627\u0644\u0627\u062e\u062a\u0644\u0627\u0641\u0627\u062a \u0627\u0644\u062a\u064a \u0644\u0627 \u062a\u062a\u062c\u0627\u0648\u0632 \u0646\u0642\u0637\u0629 \u0623\u0648 \u0646\u0642\u0637\u062a\u064a\u0646 \u0641\u064a \u0642\u0645\u0629 \u0645\u0642\u064a\u0627\u0633\u064d \u0645\u0634\u0628\u0639\u064d \u0645\u0627 \u0647\u064a \u0636\u0648\u0636\u0627\u0621 (noise)\u060c \u0648\u0644\u064a\u0633\u062a \u0625\u0634\u0627\u0631\u0629\u064b \u0630\u0627\u062a \u062f\u0644\u0627\u0644\u0629 (signal).<\/strong> \u0627\u0644\u0633\u0639\u0631 \u062c\u0632\u0621\u064c \u0645\u0646 \u0627\u0644\u062f\u0631\u062c\u0629.
\n\u0641\u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0627\u0644\u0623\u0641\u0636\u0644 \u0628\u062f\u0631\u062c\u062a\u064a\u0646 \u0644\u0643\u0646\u0647 \u0623\u063a\u0644\u0649 \u0633\u062a\u0629 \u0623\u0636\u0639\u0627\u0641 \u0644\u064a\u0633 \u0628\u0627\u0644\u0636\u0631\u0648\u0631\u0629 \u0623\u0641\u0636\u0644 \u0644\u0645\u0639\u0638\u0645 \u062d\u0627\u0644\u0627\u062a \u0627\u0644\u0627\u0633\u062a\u062e\u062f\u0627\u0645.
\n\u0648\u0644\u0647\u0630\u0627 \u0627\u0644\u0633\u0628\u0628 \u064a\u062a\u0636\u0645\u064e\u0651\u0646 \u0627\u0644\u062c\u062f\u0648\u0644 \u0623\u0639\u0644\u0627\u0647 \u0639\u0645\u0648\u062f\u064b\u0627 \u0644\u0640\u00ab\u0627\u0644\u0646\u062a\u064a\u062c\u0629 \u0644\u0643\u0644 \u062f\u0648\u0644\u0627\u0631\u00bb\u060c \u0648\u0644\u0645\u0627\u0630\u0627 \u064a\u0648\u062c\u062f \u0645\u0624\u0634\u0631 \u00ab\u0627\u0644\u0623\u062f\u0627\u0621 \u0645\u0642\u0627\u0628\u0644 \u0627\u0644\u0633\u0639\u0631\u00bb (price-performance index).<\/p>\n

\u0623\u064a \u0645\u0642\u064a\u0627\u0633 \u0623\u062f\u0627\u0621 \u064a\u062c\u0628 \u0623\u0646 \u062a\u0647\u062a\u0645 \u0628\u0647 \u0641\u0639\u0644\u064a\u064b\u0651\u0627\u061f<\/strong> \u0644\u0628\u0646\u0627\u0621 \u0648\u0643\u064a\u0644 \u0628\u0631\u0645\u062c\u064a (coding agent): \u0627\u0628\u062f\u0623 \u0623\u0648\u0644\u064b\u0627 \u0628\u0645\u062c\u0645\u0648\u0639\u0629 SWE-bench \u0627\u0644\u0645\u064f\u062d\u0642\u064e\u0651\u0642\u0629\u060c \u062b\u0645 Terminal-Bench \u062b\u0627\u0646\u064a\u064b\u0627.
\n\u062a\u0627\u0631\u064a\u062e \u0642\u0637\u0639 \u0628\u064a\u0627\u0646\u0627\u062a \u0627\u0644\u062a\u062f\u0631\u064a\u0628\u060c \u0648\u0647\u064a \u0645\u062a\u0648\u0641\u0631\u0629 \u0639\u0644\u0646\u064b\u0627\u060c \u0648\u0645\u0646 \u0627\u0644\u0645\u062d\u062a\u0645\u0644 \u0623\u0646 \u062a\u0643\u0648\u0646 \u0628\u0639\u0636 \u0647\u0630\u0647 \u0627\u0644\u062a\u0642\u0627\u0631\u064a\u0631 \u0645\u0648\u062c\u0648\u062f\u0629\u064b \u0628\u0627\u0644\u0641\u0639\u0644 \u0636\u0645\u0646 \u0628\u064a\u0627\u0646\u0627\u062a \u0627\u0644\u062a\u062f\u0631\u064a\u0628.
\n\u0627\u0644\u0633\u0628\u0628 \u0627\u0644\u0631\u0626\u064a\u0633\u064a \u0641\u064a \u0623\u0646 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0627\u0644\u0642\u062f\u064a\u0645\u0629 \u062a\u0632\u062f\u0627\u062f \u0642\u064a\u0645\u062a\u0647\u0627 \u0628\u0634\u0643\u0644 \u0645\u064f\u0636\u062e\u064e\u0651\u0645 \u0645\u0639 \u0645\u0631\u0648\u0631 \u0627\u0644\u0648\u0642\u062a\u060c \u0648\u0644\u0645\u0627\u0630\u0627 \u062a\u064f\u0635\u0645\u064e\u0651\u0645 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0627\u0644\u0623\u062d\u062f\u062b \u0644\u062a\u0643\u0648\u0646
\n\u00ab\u0645\u062d\u0635\u064e\u0651\u0646\u0629 \u0636\u062f \u062c\u0648\u062c\u0644\u00bb \u0623\u0648 \u062a\u064f\u0633\u062a\u0628\u0639\u062f \u062a\u0645\u0627\u0645\u064b\u0627.<\/p>\n

\u0627\u0646\u062a\u0628\u0647 \u0625\u0644\u0649 \u0627\u0646\u062a\u0642\u0627\u0644 \u062e\u0637 \u0627\u0644\u0646\u0647\u0627\u064a\u0629.<\/strong> \u0627\u062e\u062a\u0628\u0627\u0631 \u00abSWE-bench\u00bb \u0628\u062f\u0648\u0646 \u0639\u0644\u0627\u0645\u0629 \u00abVerified\u00bb\u060c \u0648\u0647\u0648 \u0645\u062c\u0645\u0648\u0639\u0629 \u0641\u0631\u0639\u064a\u0629 \u063a\u064a\u0631 \u0645\u0633\u0645\u064e\u0651\u0627\u0629\u060c
\n\u0639\u062f\u062f \u0645\u062e\u062a\u0644\u0641 \u0645\u0646 \u0627\u0644\u0645\u062d\u0627\u0648\u0644\u0627\u062a\u060c \u0623\u0648 \u062f\u0631\u062c\u0629 \u062a\u062d\u0642\u064e\u0651\u0642 \u0628\u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0623\u062f\u0648\u0627\u062a \u0645\u0642\u0627\u0631\u0646\u0629\u064b \u0628\u062a\u0644\u0643 \u0627\u0644\u062a\u064a \u0644\u0627 \u062a\u0633\u062a\u062e\u062f\u0645 \u0623\u062f\u0648\u0627\u062a \u2014 \u0647\u0630\u0647
\n\u0647\u064a \u0627\u0644\u0637\u0631\u0642 \u0627\u0644\u0645\u0639\u062a\u0627\u062f\u0629 \u0627\u0644\u062a\u064a \u062a\u062a\u0648\u0642\u064e\u0651\u0641 \u0628\u0647\u0627 \u0627\u0644\u0645\u0642\u0627\u0631\u0646\u0629 \u0639\u0646 \u0643\u0648\u0646\u0647\u0627 \u00ab\u0645\u0642\u0627\u0628\u0644\u0629\u064b \u0644\u0645\u0642\u0627\u0628\u0644\u0629\u00bb.<\/p>\n

\u0627\u062d\u0630\u0631 \u0645\u0646 \u0627\u0644\u062a\u0634\u0628\u0639.<\/strong> \u0639\u0646\u062f\u0645\u0627 \u064a\u062d\u0635\u0644 \u0643\u0644 \u0646\u0645\u0648\u0630\u062c\u064d \u062c\u0627\u062f\u064d\u0651 \u0639\u0644\u0649 \u062f\u0631\u062c\u0629 \u062a\u0641\u0648\u0642 \u0669\u0660 \u0641\u064a \u0627\u062e\u062a\u0628\u0627\u0631 \u0645\u0627\u060c \u0641\u0625\u0646 \u0647\u0630\u0627
\n\u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631 \u064a\u062a\u0648\u0642\u0641 \u0639\u0646 \u0627\u0644\u062a\u0645\u064a\u064a\u0632 \u0628\u064a\u0646 \u0627\u0644\u0646\u0645\u0627\u0630\u062c. \u0641\u0627\u0644\u0627\u062e\u062a\u0644\u0627\u0641\u0627\u062a \u0627\u0644\u062a\u064a \u062a\u0628\u0644\u063a \u0646\u0642\u0637\u0629 \u0623\u0648 \u0646\u0642\u0637\u062a\u064a\u0646 \u0641\u064a \u0642\u0645\u0629 \u0627\u062e\u062a\u0628\u0627\u0631\u064d \u0645\u0634\u0628\u0639\u064d
\n\u0647\u064a \u0636\u0648\u0636\u0627\u0621\u064c \u0648\u0644\u064a\u0633\u062a \u0625\u0634\u0627\u0631\u0629\u064b \u0630\u0627\u062a \u062f\u0644\u0627\u0644\u0629.<\/p>\n

\u0627\u0644\u0633\u0639\u0631 \u062c\u0632\u0621\u064c \u0645\u0646 \u0627\u0644\u062f\u0631\u062c\u0629.<\/strong> \u0646\u0645\u0648\u0630\u062c\u064c \u0623\u0641\u0636\u0644 \u0628\u062f\u0631\u062c\u062a\u064a\u0646 \u0648\u0623\u0639\u0644\u0649 \u062a\u0643\u0644\u0641\u0629\u064b \u0628\u0633\u062a\u0629 \u0623\u0636\u0639\u0627\u0641
\n\u0627\u0644\u063a\u0627\u0644\u064a \u0644\u064a\u0633 \u0628\u0627\u0644\u0636\u0631\u0648\u0631\u0629 \u0627\u0644\u0623\u0641\u0636\u0644 \u0628\u0627\u0644\u0646\u0633\u0628\u0629 \u0644\u0645\u0639\u0638\u0645 \u0623\u062d\u0645\u0627\u0644 \u0627\u0644\u0639\u0645\u0644. \u0648\u0644\u0647\u0630\u0627 \u0627\u0644\u0633\u0628\u0628 \u062a\u062a\u0636\u0645\u0646 \u0627\u0644\u062c\u062f\u0648\u0644 \u0623\u0639\u0644\u0627\u0647 \u0639\u0645\u0648\u062f \u00ab\u0627\u0644\u0646\u062a\u064a\u062c\u0629 \u0644\u0643\u0644 \u062f\u0648\u0644\u0627\u0631\u00bb\u060c
\n\u0648\u0644\u0647\u0630\u0627 \u0627\u0644\u0633\u0628\u0628 \u062a\u0648\u062c\u062f \u0645\u0624\u0634\u0631 \u0627\u0644\u0643\u0641\u0627\u0621\u0629 \u0627\u0644\u0633\u0639\u0631\u064a\u0629
\n\u0623\u064a\u0636\u064b\u0627.<\/a> \u0623\u064a \u0627\u062e\u062a\u0628\u0627\u0631 \u0645\u0639\u064a\u0627\u0631\u064a \u064a\u062c\u0628 \u0623\u0646 \u062a\u0647\u062a\u0645 \u0628\u0647 \u0641\u0639\u0644\u064a\u064b\u0651\u0627\u061f<\/p>\n

\u0628\u0646\u0627\u0621 \u0648\u0643\u064a\u0644 \u0628\u0631\u0645\u062c\u064a:<\/h2>\n

\u064a\u064f\u0641\u0636\u064e\u0651\u0644 \u0627\u062e\u062a\u0628\u0627\u0631 SWE-bench \u0627\u0644\u0645\u064f\u062d\u0642\u064e\u0651\u0642 \u0623\u0648\u0644\u064b\u0627\u060c \u062b\u0645 \u0627\u062e\u062a\u0628\u0627\u0631 Terminal-Bench \u062b\u0627\u0646\u064a\u064b\u0627.<\/strong> SWE-bench \u062a\u0645 \u0627\u0644\u062a\u062d\u0642\u0642 \u0645\u0646\u0647 \u0623\u0648\u0644\u064b\u0627\u060c \u0648Terminal-Bench \u062b\u0627\u0646\u064a\u064b\u0627.
\n\u062a\u062c\u0627\u0647\u0644 \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u062a\u0645\u0627\u0645\u064b\u0627.<\/p>\n

\u0628\u0646\u0627\u0621 \u0648\u0643\u064a\u0644\u064d \u0644\u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0627\u0644\u062d\u0627\u0633\u0648\u0628 \u0623\u0648 \u0633\u0637\u062d \u0627\u0644\u0645\u0643\u062a\u0628:<\/strong> OSWorld\u060c \u062b\u0645 Terminal-Bench.
\n\u062a\u0648\u0642\u064e\u0651\u0639 \u0623\u0646 \u062a\u0643\u0648\u0646 \u0627\u0644\u0623\u0631\u0642\u0627\u0645 \u0627\u0644\u0645\u0637\u0644\u0642\u0629 \u0645\u0646\u062e\u0641\u0636\u0629\u064b \u0646\u0633\u0628\u064a\u064b\u0651\u0627.<\/p>\n

\u0627\u0644\u0625\u062c\u0627\u0628\u0629 \u0639\u0646 \u0627\u0644\u0623\u0633\u0626\u0644\u0629 \u0627\u0644\u062a\u0642\u0646\u064a\u0629 \u0623\u0648 \u0627\u0644\u0639\u0644\u0645\u064a\u0629:<\/strong> GPQA\u060c \u0648\u0627\u0645\u062a\u062d\u0627\u0646 \u0627\u0644\u0628\u0634\u0631\u064a\u0629 \u0627\u0644\u0623\u062e\u064a\u0631 (Humanity's Last Exam) \u0625\u0630\u0627 \u0643\u0627\u0646\u062a
\n\u0623\u0633\u0626\u0644\u062a\u0643 \u0628\u0627\u0644\u0641\u0639\u0644 \u0639\u0644\u0649 \u0645\u0633\u062a\u0648\u0649 \u0627\u0644\u062e\u0628\u0631\u0627\u0621 \u0627\u0644\u0645\u062a\u062e\u0635\u0635\u064a\u0646.<\/p>\n

\u0627\u0644\u0648\u0643\u064a\u0644 \u0627\u0644\u0639\u0627\u0645 \u0623\u0648 \u0631\u0648\u0628\u0648\u062a \u0627\u0644\u062f\u0631\u062f\u0634\u0629:<\/strong> \u0627\u0644\u0645\u0624\u0634\u0631 \u0627\u0644\u0645\u0631\u0643\u0628 \u064a\u0648\u0641\u0650\u0651\u0631 \u0645\u0639\u0644\u0648\u0645\u0627\u062a\u064d \u0623\u0643\u062b\u0631 \u0641\u0627\u0626\u062f\u0629\u064b \u0645\u0646 \u0623\u064a \u0627\u062e\u062a\u0628\u0627\u0631\u064d \u0648\u0627\u062d\u062f\u064d\u060c \u0644\u0623\u0646\u064e\u0651\u0647 \u0644\u0627 \u064a\u0648\u062c\u062f \u0627\u062e\u062a\u0628\u0627\u0631\u064c \u0648\u062d\u064a\u062f\u064c \u064a\u064f\u0645\u0627\u062b\u0644 \u0627\u0644\u062a\u0646\u0648\u064f\u0651\u0639 \u0627\u0644\u0648\u0627\u0633\u0639 \u0641\u064a \u0627\u0644\u0623\u0633\u0626\u0644\u0629 \u0627\u0644\u062a\u064a \u064a\u0637\u0631\u062d\u0647\u0627 \u0627\u0644\u0645\u0633\u062a\u062e\u062f\u0645\u0648\u0646 \u0627\u0644\u0641\u0639\u0644\u064a\u0648\u0646.
\n\u0627\u062e\u062a\u0628\u0627\u0631\u064c \u0648\u0627\u062d\u062f\u064c \u0641\u0642\u0637\u060c \u0644\u0623\u0646\u064e\u0651 \u0623\u064a\u064e\u0651 \u0627\u062e\u062a\u0628\u0627\u0631 \u0645\u0639\u064a\u0646 \u0644\u0627 \u064a\u0639\u0643\u0633 \u0627\u0644\u062a\u0646\u0648\u064f\u0651\u0639 \u0627\u0644\u0648\u0627\u0633\u0639 \u0641\u064a \u0627\u0644\u0645\u0647\u0627\u0645 \u0627\u0644\u062a\u064a \u064a\u0637\u0644\u0628\u0647\u0627 \u0627\u0644\u0645\u0633\u062a\u062e\u062f\u0645\u0648\u0646 \u0627\u0644\u062d\u0642\u064a\u0642\u064a\u0648\u0646 \u0641\u0639\u0644\u064b\u0627.<\/p>\n

\u0627\u062e\u062a\u064a\u0627\u0631 \u0646\u0645\u0648\u0630\u062c \u0635\u063a\u064a\u0631 \u0644\u0644\u062a\u0634\u063a\u064a\u0644 \u0627\u0644\u0630\u0627\u062a\u064a (Self-hosting):<\/strong> \u062a\u0644\u0639\u0628 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u062f\u0648\u0631\u064b\u0627 \u0623\u0642\u0644 \u0623\u0647\u0645\u064a\u0629\u064b \u0645\u0642\u0627\u0631\u0646\u0629\u064b \u0628\u0645\u062f\u0649 \u0645\u0644\u0627\u0621\u0645\u0629 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0644\u0627\u062d\u062a\u064a\u0627\u062c\u0627\u062a\u0643. \u062a\u062d\u0642\u064e\u0651\u0642 \u0623\u0648\u0644\u064b\u0627 \u0645\u0646
\n\u0645\u0627 \u064a\u0645\u0643\u0646\u0647 \u0627\u0644\u062a\u0634\u063a\u064a\u0644 \u0641\u0639\u0644\u064a\u064b\u0651\u0627 \u0639\u0644\u0649 \u0623\u062c\u0647\u0632\u062a\u0643 \u0627\u0644\u0645\u0627\u062f\u064a\u0629\u060c \u062b\u0645 \u0642\u0627\u0631\u0646
\n\u062d\u0627\u0633\u0628\u0629 \u0630\u0627\u0643\u0631\u0629 \u0627\u0644\u0641\u064a\u062f\u064a\u0648 (VRAM)<\/a> \u0646\u062a\u0627\u0626\u062c \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0641\u0642\u0637 \u0628\u064a\u0646 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u062a\u064a \u062a\u062a\u0648\u0627\u0641\u0642 \u0645\u0639 \u0645\u0648\u0627\u0635\u0641\u0627\u062a \u0623\u062c\u0647\u0632\u062a\u0643.
\n\u0627\u0644\u0646\u062a\u0627\u0626\u062c \u0627\u0644\u0645\u0646\u0634\u0648\u0631\u0629 \u0644\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0641\u064a \u0642\u0627\u0639\u062f\u062a\u0646\u0627 \u0627\u0644\u0628\u064a\u0627\u0646\u0627\u062a<\/p>\n

\u0627\u0644\u0646\u062a\u0627\u0626\u062c \u0627\u0644\u0645\u0646\u0634\u0648\u0631\u0629<\/h2>\n
\n \n \n \n \n \n \n
\u0627\u0644\u0646\u0645\u0648\u0630\u062c<\/th>\u0645\u0624\u0634\u0631 \u0627\u0644\u0630\u0643\u0627\u0621 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a \u0644\u0634\u0631\u0643\u0629 Artificial Analysis: \u0665\u0669.<\/th><\/tr><\/thead>\n
\u062c\u064a \u0628\u064a \u062a\u064a-\u0665.\u0666 \u0633\u0648\u0644<\/a><\/td>\n SWE-Bench Verified: \u0668\u0665,\u0662\u066a\u061b SWE-Bench Pro: \u0666\u0663,\u0662\u066a\u061b Terminal-Bench 2.1: \u0668\u0660,\u0664\u066a\u061b OSWorld-Verified: \u0668\u0661,\u0662\u066a.<\/td>\n <\/tr>\n
\u0643\u0644\u0627\u0648\u062f \u0633\u0648\u0646\u064a\u062a \u0665<\/a><\/td>\n \u0645\u0624\u0634\u0631 \u0627\u0644\u0630\u0643\u0627\u0621 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a \u0644\u0634\u0631\u0643\u0629 Artificial Analysis: \u0666\u0661 (\u0645\u0631\u062a\u0628\u0629 \u0631\u0642\u0645 \u0661 \u0645\u0646 \u0623\u0635\u0644 \u0661\u0667\u0660 \u0646\u0645\u0648\u0630\u062c\u064b\u0627). \u064a\u0641\u0648\u0642 \u0623\u062f\u0627\u0621 Opus 4.8 \u0628\u0623\u0643\u062b\u0631 \u0645\u0646 \u0627\u0644\u0636\u0639\u0641 \u0641\u064a \u0627\u062e\u062a\u0628\u0627\u0631 Frontier-Bench \u0627\u0644\u0625\u0635\u062f\u0627\u0631 \u0660.\u0661\u061b \u0648\u064a\u0628\u0644\u063a \u0646\u062d\u0648 \u062b\u0644\u0627\u062b\u0629 \u0623\u0636\u0639\u0627\u0641 \u0623\u062f\u0627\u0621 \u0623\u0641\u0636\u0644 \u0646\u0645\u0648\u0630\u062c \u062a\u0627\u0644\u064a \u0641\u064a \u0627\u062e\u062a\u0628\u0627\u0631 ARC-AGI 3.<\/td>\n <\/tr>\n
\u0643\u0644\u0627\u0648\u062f \u0623\u0648\u0628\u0633 \u0665<\/a><\/td>\n OCRBench V2: \u0666\u0667,\u0660\u0664 | Video-MME: \u0667\u0662,\u0662 | OSWorld: \u0664\u0667,\u0664 | Speech IF: \u0668\u0669,\u0663\u0669<\/td>\n <\/tr>\n
\u0625\u0646\u0641\u064a\u062f\u064a\u0627 \u0646\u064a\u0645\u0648\u062a\u0631\u0648\u0646 \u0663 \u0646\u0627\u0646\u0648 \u0623\u0648\u0645\u0646\u064a<\/a><\/td>\n \u064a\u062a\u0645 \u0625\u062f\u0631\u0627\u062c \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0641\u0642\u0637 \u0625\u0630\u0627 \u0646\u0634\u0631 \u0645\u0637\u0648\u0650\u0651\u0631\u0648\u0647\u0627 \u0623\u0631\u0642\u0627\u0645\u064b\u0627 \u0645\u062d\u062f\u062f\u0629\u064b \u0628\u0648\u0636\u0648\u062d. \u0648\u0628\u0627\u0644\u062a\u0627\u0644\u064a \u0641\u0625\u0646 \u063a\u064a\u0627\u0628 \u0646\u0645\u0648\u0630\u062c\u064d \u0645\u0627 \u0647\u0646\u0627 \u0644\u0627 \u064a\u0639\u0646\u064a \u0623\u0646 \u0623\u062f\u0627\u0624\u0647 \u0636\u0639\u064a\u0641\u064c\u060c \u0628\u0644 \u064a\u0639\u0646\u064a \u0641\u0642\u0637 \u0623\u0646\u0646\u0627 \u0644\u0645 \u0646\u062a\u062d\u0642\u0642 \u0628\u0639\u062f\u064f \u0645\u0646 \u0648\u062c\u0648\u062f \u0631\u0642\u0645 \u0645\u0646\u0634\u0648\u0631\u064d \u0644\u0647 \u2014 \u0648\u0647\u0630\u0647 \u0627\u0644\u0645\u0639\u0644\u0648\u0645\u0629 \u0641\u064a \u062d\u062f \u0630\u0627\u062a\u0647\u0627 \u0630\u0627\u062a \u0642\u064a\u0645\u0629\u064d \u0643\u0628\u064a\u0631\u0629\u064d \u0639\u0646\u062f\u0645\u0627 \u064a\u0631\u0648\u0650\u0651\u062c \u0627\u0644\u0645\u0648\u0631\u0650\u0651\u062f \u0644\u0627\u062f\u0639\u0627\u0621\u064d \u0645\u0628\u0646\u064a\u064d\u0651 \u0639\u0644\u0649 \u0646\u062a\u064a\u062c\u0629 \u0627\u062e\u062a\u0628\u0627\u0631\u064d \u0645\u0627.<\/td>\n <\/tr>\n <\/tbody>\n <\/table>\n

\u0645\u0627 \u0647\u0648 \u0627\u062e\u062a\u0628\u0627\u0631 \u0627\u0644\u0630\u0643\u0627\u0621 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a (AI Benchmark)\u061f<\/p>\n<\/div>\n\n

\u0627\u0644\u0623\u0633\u0626\u0644\u0629 \u0627\u0644\u0634\u0627\u0626\u0639\u0629<\/h2>\n
\n
\n\u0645\u062c\u0645\u0648\u0639\u0629 \u062b\u0627\u0628\u062a\u0629 \u0645\u0646 \u0627\u0644\u0645\u0647\u0627\u0645 \u0630\u0627\u062a \u0637\u0631\u064a\u0642\u0629 \u062a\u0642\u064a\u064a\u0645 \u0645\u064f\u0639\u0631\u064e\u0651\u0641\u0629 \u0628\u0648\u0636\u0648\u062d\u060c \u062a\u064f\u0633\u062a\u062e\u062f\u0645 \u0644\u0645\u0642\u0627\u0631\u0646\u0629 \u0623\u062f\u0627\u0621 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0645\u062e\u062a\u0644\u0641\u0629 \u0639\u0644\u0649 \u0645\u062f\u062e\u0644\u0627\u062a\u064d \u0645\u062a\u0637\u0627\u0628\u0642\u0629. \u0648\u062a\u062a\u0631\u0627\u0648\u062d \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0645\u0646 \u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0627\u0644\u0645\u0639\u0631\u0641\u0629 \u0645\u062a\u0639\u062f\u062f\u0629 \u0627\u0644\u062e\u064a\u0627\u0631\u0627\u062a \u0645\u062b\u0644 MMLU\u060c \u0625\u0644\u0649 \u0627\u0644\u0645\u0647\u0627\u0645 \u0627\u0644\u0648\u0643\u064a\u0644\u064a\u0629 (Agentic Tasks) \u0645\u062b\u0644 \u062d\u0644 \u0645\u0634\u0643\u0644\u0627\u062a \u062d\u0642\u064a\u0642\u064a\u0629 \u0639\u0644\u0649 GitHub \u0636\u0645\u0646 \u0627\u062e\u062a\u0628\u0627\u0631 SWE-Bench\u060c \u0648\u0643\u0644 \u0645\u0646\u0647\u0627 \u064a\u0642\u064a\u0633 \u062c\u0627\u0646\u0628\u064b\u0627 \u0636\u064a\u0642\u064b\u0627 \u062c\u062f\u064b\u0651\u0627 \u0645\u0646 \u0627\u0644\u0642\u062f\u0631\u0627\u062a\u060c \u0648\u0644\u064a\u0633 \u0627\u0644\u062c\u0648\u062f\u0629 \u0627\u0644\u0634\u0627\u0645\u0644\u0629.<\/summary>\n

\u0623\u064a \u0627\u062e\u062a\u0628\u0627\u0631 \u0630\u0643\u0627\u0621 \u0627\u0635\u0637\u0646\u0627\u0639\u064a \u0647\u0648 \u0627\u0644\u0623\u0643\u062b\u0631 \u0645\u0648\u062b\u0648\u0642\u064a\u0629\u064b\u061f
\n\u0644\u0627 \u064a\u0648\u062c\u062f \u0627\u062e\u062a\u0628\u0627\u0631\u064c \u0648\u0627\u062d\u062f\u064c \u064a\u064f\u0639\u062a\u0628\u0631 \u0627\u0644\u0623\u0643\u062b\u0631 \u0645\u0648\u062b\u0648\u0642\u064a\u0629\u064b. \u0641\u0627\u0644\u0645\u0624\u0634\u0631\u0627\u062a \u0627\u0644\u0645\u0631\u0643\u0628\u0629 \u0647\u064a \u0627\u0644\u0631\u0642\u0645 \u0627\u0644\u0631\u0626\u064a\u0633\u064a \u0627\u0644\u0623\u0643\u062b\u0631 \u0645\u0648\u062b\u0648\u0642\u064a\u0629\u064b\u060c \u0644\u0623\u0646 \u062f\u0645\u062c \u0639\u062f\u0629 \u062a\u0642\u064a\u064a\u0645\u0627\u062a\u064d \u064a\u0635\u0639\u0628 \u0627\u0644\u062a\u0644\u0627\u0639\u0628 \u0628\u0647 \u0645\u0642\u0627\u0631\u0646\u0629\u064b \u0628\u0627\u0644\u062a\u0631\u0643\u064a\u0632 \u0639\u0644\u0649 \u062a\u062d\u0633\u064a\u0646 \u0627\u0644\u0623\u062f\u0627\u0621 \u0641\u064a \u0627\u062e\u062a\u0628\u0627\u0631\u064d \u0648\u0627\u062d\u062f\u064d \u0641\u0642\u0637. \u0623\u0645\u0627 \u0628\u0627\u0644\u0646\u0633\u0628\u0629 \u0644\u0642\u0631\u0627\u0631\u064d \u0645\u0639\u064a\u064e\u0651\u0646\u064d\u060c \u0641\u0625\u0646 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631 \u0627\u0644\u0623\u0643\u062b\u0631 \u0645\u0648\u062b\u0648\u0642\u064a\u0629\u064b \u0647\u0648 \u0627\u0644\u0630\u064a \u064a\u0634\u0628\u0647 \u0625\u0644\u0649 \u0623\u0642\u0635\u0649 \u062d\u062f\u064d\u0651 \u0645\u0645\u0643\u0646\u064d \u0637\u0628\u064a\u0639\u0629 \u0627\u0644\u0645\u0647\u0645\u0629 \u0627\u0644\u062a\u064a \u0633\u062a\u064f\u0646\u0627\u0637 \u0628\u0647 \u0641\u064a \u0627\u0644\u0648\u0627\u0642\u0639 \u2014 \u0643\u0623\u0646 \u062a\u062e\u062a\u0627\u0631 SWE-Bench Verified \u0644\u0644\u0648\u0643\u0627\u0644\u0627\u062a \u0627\u0644\u0628\u0631\u0645\u062c\u064a\u0629\u060c \u0623\u0648 OSWorld \u0644\u0645\u0647\u0627\u0645 \u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0627\u0644\u062d\u0627\u0633\u0648\u0628\u060c \u0623\u0648 GPQA \u0644\u0644\u062a\u0641\u0643\u064a\u0631 \u0627\u0644\u062a\u0642\u0646\u064a.
\n\u0645\u0627 \u0627\u0644\u062f\u0631\u062c\u0629 \u0627\u0644\u062c\u064a\u062f\u0629 \u0641\u064a \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0639\u0627\u0645 \u0662\u0660\u0662\u0666\u061f
\n\u0625\u0646 \u062f\u0631\u062c\u0627\u062a \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0642\u062f \u0627\u0642\u062a\u0631\u0628\u062a \u0645\u0646 \u0627\u0644\u062a\u0634\u0628\u0639 \u0627\u0644\u0643\u0627\u0645\u0644 \u0628\u064a\u0646 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0631\u0627\u0626\u062f\u0629\u060c \u0648\u0628\u0627\u0644\u062a\u0627\u0644\u064a \u0644\u0645 \u062a\u0639\u062f \u0627\u0644\u062f\u0631\u062c\u0629 \u0627\u0644\u0639\u0627\u0644\u064a\u0629 \u062a\u064f\u0645\u064a\u0650\u0651\u0632 \u0628\u064a\u0646 \u0647\u0630\u0647 \u0627\u0644\u0646\u0645\u0627\u0630\u062c. \u0648\u0645\u0639 \u0630\u0644\u0643\u060c \u0644\u0627 \u064a\u0632\u0627\u0644 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631 \u0645\u0641\u064a\u062f\u064b\u0627 \u0641\u064a \u062a\u0642\u064a\u064a\u0645 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0645\u0641\u062a\u0648\u062d\u0629 \u0627\u0644\u0623\u0635\u063a\u0631\u060c \u062d\u064a\u062b \u0644\u0627 \u064a\u0632\u0627\u0644 \u0627\u0644\u062a\u0628\u0627\u064a\u0646 \u0641\u064a \u0627\u0644\u0623\u062f\u0627\u0621 \u0648\u0627\u0633\u0639\u064b\u0627 \u0646\u0633\u0628\u064a\u064b\u0651\u0627. \u0641\u0625\u0630\u0627 \u0643\u0646\u062a \u062a\u0642\u0627\u0631\u0646 \u0628\u064a\u0646 \u0646\u0645\u0648\u0630\u062c\u064a\u0646 \u0631\u0627\u0626\u062f\u064a\u0646\u060c \u0641\u0644\u0646 \u064a\u064f\u0638\u0647\u0631 \u0644\u0643 \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0641\u0631\u0642\u064b\u0627 \u0628\u064a\u0646\u0647\u0645\u0627\u060c \u0648\u0644\u0646 \u062a\u062d\u062a\u0627\u062c \u062d\u064a\u0646\u0647\u0627 \u0625\u0644\u0649 \u0627\u0644\u0627\u0639\u062a\u0645\u0627\u062f \u0639\u0644\u0649 GPQA \u0623\u0648 \u0645\u0624\u0634\u0631\u064d \u0645\u0631\u0643\u0628\u064d \u0628\u062f\u0644\u064b\u0627 \u0645\u0646\u0647.<\/p>\n<\/details>\n

\n\u0647\u0644 \u064a\u0645\u0643\u0646 \u0627\u0644\u062a\u0644\u0627\u0639\u0628 \u0628\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0627\u0644\u0630\u0643\u0627\u0621 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a\u061f<\/summary>\n

\u0646\u0639\u0645\u060c \u0648\u0628\u0637\u0631\u064a\u0642\u062a\u064a\u0646. \u0627\u0644\u0623\u0648\u0644\u0649: \u0623\u0646 \u062a\u062a\u0636\u0645\u064e\u0651\u0646 \u0628\u064a\u0627\u0646\u0627\u062a \u0627\u0644\u062a\u062f\u0631\u064a\u0628 \u0623\u0633\u0626\u0644\u0629 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631 \u0646\u0641\u0633\u0647\u060c \u0645\u0645\u0627 \u064a\u0631\u0641\u0639 \u0627\u0644\u062f\u0631\u062c\u0627\u062a \u062f\u0648\u0646 \u062a\u062d\u0642\u064a\u0642 \u0623\u064a \u062a\u062d\u0633\u064f\u0651\u0646 \u062d\u0642\u064a\u0642\u064a \u0641\u064a \u0627\u0644\u0642\u062f\u0631\u0627\u062a \u2014 \u0648\u0644\u0630\u0644\u0643 \u0635\u064f\u0645\u0650\u0651\u0645\u062a \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0627\u0644\u062c\u062f\u064a\u062f\u0629 \u0644\u0645\u0642\u0627\u0648\u0645\u0629 \u0639\u0645\u0644\u064a\u0627\u062a \u0627\u0644\u0628\u062d\u062b \u0648\u0627\u0644\u062d\u0641\u0638 \u0627\u0644\u0622\u0644\u064a. \u0648\u0627\u0644\u062b\u0627\u0646\u064a\u0629: \u0627\u062e\u062a\u064a\u0627\u0631 \u0638\u0631\u0648\u0641 \u0627\u0644\u062a\u0642\u064a\u064a\u0645 \u0628\u0637\u0631\u064a\u0642\u0629 \u0645\u064f\u0641\u0636\u064e\u0651\u0644\u0629\u060c \u0645\u062b\u0644 \u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0645\u062c\u0645\u0648\u0639\u0629 \u0641\u0631\u0639\u064a\u0629 \u0645\u062e\u062a\u0644\u0641\u0629 \u0645\u0646 \u0627\u0644\u0623\u0633\u0626\u0644\u0629\u060c \u0623\u0648 \u0632\u064a\u0627\u062f\u0629 \u0639\u062f\u062f \u0627\u0644\u0645\u062d\u0627\u0648\u0644\u0627\u062a\u060c \u0623\u0648 \u0627\u0644\u0633\u0645\u0627\u062d \u0628\u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0623\u062f\u0648\u0627\u062a \u0645\u0633\u0627\u0639\u062f\u0629 \u0628\u064a\u0646\u0645\u0627 \u0644\u0627 \u064a\u064f\u0633\u0645\u062d \u0628\u0647\u0627 \u0644\u0644\u0646\u0645\u0648\u0630\u062c \u0627\u0644\u0622\u062e\u0631. \u0648\u062a\u062e\u0641\u0650\u0651\u0641 \u0627\u0644\u0644\u0648\u062d\u0627\u062a \u0627\u0644\u0642\u064a\u0627\u062f\u064a\u0629 \u0627\u0644\u0645\u0633\u062a\u0642\u0644\u0629 \u0648\u0627\u0644\u0645\u0624\u0634\u0631\u0627\u062a \u0627\u0644\u0645\u0631\u0643\u0628\u0629 \u0645\u0646 \u0647\u0627\u062a\u064a\u0646 \u0627\u0644\u0645\u0634\u0643\u0644\u062a\u064a\u0646 \u0645\u0639\u064b\u0627.
\n\u0647\u0644 \u062a\u062a\u0646\u0628\u064e\u0651\u0623 \u062f\u0631\u062c\u0627\u062a \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0628\u0627\u0644\u0623\u062f\u0627\u0621 \u0627\u0644\u0641\u0639\u0644\u064a \u0641\u064a \u0627\u0644\u0639\u0627\u0644\u0645 \u0627\u0644\u062d\u0642\u064a\u0642\u064a\u061f
\n\u062c\u0632\u0626\u064a\u064b\u0651\u0627 \u0641\u0642\u0637. \u0641\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0627\u0644\u0645\u0647\u0627\u0645 \u0627\u0644\u0623\u0635\u064a\u0644\u0629 (Task-authentic benchmarks)\u060c \u0645\u062b\u0644 SWE-Bench Verified \u0648OSWorld\u060c \u062a\u062a\u0646\u0628\u064e\u0651\u0623 \u0628\u0634\u0643\u0644\u064d \u0645\u0639\u0642\u0648\u0644\u064d \u062c\u064a\u062f\u064d \u0644\u0623\u0646 \u0627\u0644\u0645\u0647\u0645\u0629 \u0646\u0641\u0633\u0647\u0627 \u0647\u064a \u0627\u0644\u0645\u0647\u0645\u0629 \u0627\u0644\u0648\u0627\u0642\u0639\u064a\u0629. \u0623\u0645\u0627 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631\u0627\u062a \u0627\u0644\u0623\u0643\u0627\u062f\u064a\u0645\u064a\u0629 \u0645\u062a\u0639\u062f\u062f\u0629 \u0627\u0644\u062e\u064a\u0627\u0631\u0627\u062a \u0641\u062a\u0646\u0628\u0650\u0651\u0626 \u0628\u0634\u0643\u0644\u064d \u0636\u0639\u064a\u0641\u064d \u062c\u062f\u064b\u0651\u0627\u060c \u0644\u0623\u0646 \u0627\u0644\u0625\u062c\u0627\u0628\u0629 \u0639\u0646 \u0623\u0633\u0626\u0644\u0629 \u0627\u0644\u0627\u0645\u062a\u062d\u0627\u0646\u0627\u062a \u0644\u064a\u0633\u062a \u0627\u0644\u0645\u0647\u0645\u0629 \u0627\u0644\u062a\u064a \u062a\u0624\u062f\u064a\u0647\u0627 \u0645\u0639\u0638\u0645 \u0627\u0644\u062a\u0637\u0628\u064a\u0642\u0627\u062a \u0641\u064a \u0627\u0644\u0648\u0627\u0642\u0639. \u0648\u0623\u0642\u0648\u0649 \u0645\u0624\u0634\u0631 \u062a\u0646\u0628\u0624\u064a\u064d\u0651 \u0644\u0627 \u064a\u0632\u0627\u0644 \u0647\u0648 \u0627\u0644\u062a\u0642\u064a\u064a\u0645 \u0627\u0644\u0630\u064a \u062a\u0642\u0648\u0645 \u0628\u0628\u0646\u0627\u0626\u0647 \u0628\u0646\u0641\u0633\u0643 \u0639\u0644\u0649 \u0628\u064a\u0627\u0646\u0627\u062a\u0643 \u0627\u0644\u062e\u0627\u0635\u0629.
\n\u0623\u064a \u0646\u0645\u0648\u0630\u062c \u0645\u0641\u062a\u0648\u062d \u0627\u0644\u0648\u0632\u0646 (Open-weight model) \u064a\u062d\u0642\u0642 \u0623\u0639\u0644\u0649 \u062f\u0631\u062c\u0629\u061f<\/p>\n<\/details>\n

\n\u0646\u0645\u0648\u0630\u062c Kimi K3 \u0628\u0646\u062a\u064a\u062c\u0629 \u0665\u0667 \u0641\u064a \u0645\u0624\u0634\u0631 \u0627\u0644\u0630\u0643\u0627\u0621 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a \u0644\u0634\u0631\u0643\u0629 Artificial Analysis\u060c \u0648\u0647\u0648 \u0623\u0641\u0636\u0644 \u0646\u062a\u064a\u062c\u0629 \u0645\u0633\u062c\u064e\u0651\u0644\u0629 \u062d\u062a\u0649 \u0627\u0644\u0622\u0646 \u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0648\u0632\u0646 \u0627\u0644\u0645\u0641\u062a\u0648\u062d\u060c \u0648\u064a\u0628\u062a\u0639\u062f \u0628\u0641\u0627\u0631\u0642 \u0623\u0631\u0628\u0639 \u0646\u0642\u0627\u0637 \u0641\u0642\u0637 \u0639\u0646 \u0623\u0641\u0636\u0644 \u0646\u0645\u0648\u0630\u062c \u0645\u063a\u0644\u0642. \u0648\u062a\u0634\u064a\u0631 \u0627\u0644\u062c\u062f\u0648\u0644 \u0627\u0644\u0642\u064a\u0627\u062f\u064a \u0623\u0639\u0644\u0627\u0647 \u0625\u0644\u0649 \u062c\u0645\u064a\u0639 \u0627\u0644\u0625\u062f\u062e\u0627\u0644\u0627\u062a \u0627\u0644\u062e\u0627\u0635\u0629 \u0628\u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0645\u0641\u062a\u0648\u062d\u0629 \u0627\u0644\u0648\u0632\u0646\u060c \u0648\u064a\u0645\u0643\u0646 \u0641\u0631\u0632 \u0627\u0644\u0645\u062c\u0645\u0648\u0639\u0629 \u0627\u0644\u0643\u0627\u0645\u0644\u0629 \u0641\u064a \u0627\u0644\u0635\u0641\u062d\u0629.<\/summary>\n

\u0644\u0642\u062f \u0628\u0644\u063a\u062a \u062f\u0631\u062c\u0627\u062a \u0646\u0645\u0648\u0630\u062c MMLU \u062d\u062f \u0627\u0644\u062a\u0634\u0628\u0639 \u0625\u0644\u0649 \u062d\u062f\u064d\u0651 \u0643\u0628\u064a\u0631 \u0628\u064a\u0646 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0631\u0627\u0626\u062f\u0629\u060c \u0648\u0628\u0627\u0644\u062a\u0627\u0644\u064a \u0644\u0645 \u062a\u0639\u062f \u0627\u0644\u062f\u0631\u062c\u0629 \u0627\u0644\u0639\u0627\u0644\u064a\u0629 \u062a\u064f\u0645\u064a\u0650\u0651\u0632 \u0628\u064a\u0646\u0647\u0627.
\n\u0648\u0645\u0639 \u0630\u0644\u0643\u060c \u0644\u0627 \u064a\u0632\u0627\u0644 \u0647\u0630\u0627 \u0627\u0644\u0627\u062e\u062a\u0628\u0627\u0631 \u0645\u0641\u064a\u062f\u064b\u0627 \u0641\u064a \u062a\u0642\u064a\u064a\u0645 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0645\u0641\u062a\u0648\u062d\u0629 \u0627\u0644\u0623\u0635\u063a\u0631 \u062d\u062c\u0645\u064b\u0627\u060c \u062d\u064a\u062b \u0644\u0627 \u064a\u0632\u0627\u0644 \u0627\u0644\u062a\u0628\u0627\u064a\u0646 \u0641\u064a \u0627\u0644\u0623\u062f\u0627\u0621 \u0648\u0627\u0633\u0639\u064b\u0627. \u0641\u0625\u0630\u0627 \u0643\u0646\u062a\u064e
\n\u062a\u064f\u0642\u0627\u0631\u0646 \u0628\u064a\u0646 \u0646\u0645\u0648\u0630\u062c\u064a\u0646 \u0631\u0627\u0626\u062f\u064a\u0646\u060c \u0641\u0625\u0646 \u0627\u062e\u062a\u0628\u0627\u0631 MMLU \u0644\u0646 \u064a\u064f\u0638\u0647\u0631 \u0641\u0631\u0642\u064b\u0627 \u0628\u064a\u0646\u0647\u0645\u0627\u060c \u0648\u0644\u0630\u0644\u0643 \u064a\u064f\u0641\u0636\u064e\u0651\u0644 \u0627\u0644\u0627\u0639\u062a\u0645\u0627\u062f \u0639\u0644\u0649 \u0627\u062e\u062a\u0628\u0627\u0631 GPQA \u0623\u0648 \u0645\u0624\u0634\u0631 \u0645\u0631\u0643\u064e\u0651\u0628.
\n\u0633\u064a\u0641\u0639\u0644.<\/p>\n<\/details>\n

\n\u0647\u0644 \u064a\u0645\u0643\u0646 \u0627\u0644\u062a\u0644\u0627\u0639\u0628 \u0628\u0646\u062a\u0627\u0626\u062c \u0645\u0642\u0627\u064a\u064a\u0633 \u0623\u062f\u0627\u0621 \u0627\u0644\u0630\u0643\u0627\u0621 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a\u061f<\/summary>\n

\u0646\u0639\u0645\u060c \u0648\u0628\u0637\u0631\u064a\u0642\u062a\u064a\u0646. \u0641\u0642\u062f \u062a\u062a\u0636\u0645\u0651\u0646 \u0628\u064a\u0627\u0646\u0627\u062a \u0627\u0644\u062a\u062f\u0631\u064a\u0628 \u0623\u0633\u0626\u0644\u0629 \u0627\u0644\u0645\u0642\u064a\u0627\u0633 \u0646\u0641\u0633\u0647\u060c \u0645\u0627 \u064a\u0624\u062f\u064a \u0625\u0644\u0649 \u062a\u0636\u062e\u064a\u0645 \u0627\u0644\u0646\u062a\u0627\u0626\u062c \u062f\u0648\u0646 \u0623\u064a \u062a\u062d\u0633\u064f\u0651\u0646\u064d \u0641\u0639\u0644\u064a\u064d\u0651 \u0641\u064a \u0627\u0644\u0642\u062f\u0631\u0627\u062a \u2014 \u0648\u0644\u0630\u0644\u0643 \u0635\u064f\u0645\u0650\u0651\u0645\u062a \u0627\u0644\u0645\u0642\u0627\u064a\u064a\u0633 \u0627\u0644\u0623\u062d\u062f\u062b \u0644\u0645\u0642\u0627\u0648\u0645\u0629 \u0627\u0644\u0628\u062d\u062b \u0648\u0627\u0644\u062d\u0641\u0638 \u0627\u0644\u0622\u0644\u064a. \u0643\u0645\u0627 \u064a\u0645\u0643\u0646 \u0627\u062e\u062a\u064a\u0627\u0631 \u0638\u0631\u0648\u0641 \u0627\u0644\u062a\u0642\u064a\u064a\u0645 \u0628\u0637\u0631\u064a\u0642\u0629 \u0645\u064f\u0641\u0636\u064e\u0651\u0644\u0629: \u0645\u062b\u0644 \u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0645\u062c\u0645\u0648\u0639\u0629 \u062c\u0632\u0626\u064a\u0629 \u0645\u062e\u062a\u0644\u0641\u0629\u060c \u0623\u0648 \u0632\u064a\u0627\u062f\u0629 \u0639\u062f\u062f \u0627\u0644\u0645\u062d\u0627\u0648\u0644\u0627\u062a\u060c \u0623\u0648 \u0627\u0644\u0633\u0645\u0627\u062d \u0628\u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0623\u062f\u0648\u0627\u062a \u0645\u0639\u064a\u0646\u0629 \u0639\u0646\u062f \u0627\u0644\u0645\u0642\u0627\u0631\u0646\u0629 \u0628\u0646\u0645\u0648\u0630\u062c \u0644\u0627 \u064a\u0633\u062a\u062e\u062f\u0645\u0647\u0627. \u0648\u062a\u062e\u0641\u0650\u0651\u0641 \u0627\u0644\u0644\u0648\u062d\u0627\u062a \u0627\u0644\u0642\u064a\u0627\u062f\u064a\u0629 \u0627\u0644\u0645\u0633\u062a\u0642\u0644\u0629 \u0627\u0644\u062a\u064a \u062a\u064f\u062f\u0627\u0631 \u0628\u0634\u0643\u0644 \u0645\u0646\u0641\u0635\u0644 \u0648\u0627\u0644\u0645\u0624\u0634\u0631\u0627\u062a \u0627\u0644\u0645\u0631\u0643\u064e\u0651\u0628\u0629 \u0645\u0646 \u0643\u0644\u0627 \u0627\u0644\u0646\u0648\u0639\u064a\u0646 \u0645\u0646 \u0627\u0644\u062a\u062d\u064a\u064f\u0651\u0632.
\n\u0623\u064a \u0645\u0643\u0627\u0633\u0628 \u062d\u0642\u064a\u0642\u064a\u0629 \u0641\u064a \u0627\u0644\u0642\u062f\u0631\u0627\u062a \u2014 \u0648\u0644\u0630\u0644\u0643 \u0635\u064f\u0645\u0650\u0651\u0645\u062a \u0645\u0642\u0627\u064a\u064a\u0633 \u0627\u0644\u0623\u062f\u0627\u0621 \u0627\u0644\u0623\u062d\u062f\u062b \u0644\u0645\u0642\u0627\u0648\u0645\u0629 \u0639\u0645\u0644\u064a\u0627\u062a \u0627\u0644\u0628\u062d\u062b
\n\u0648\u0627\u0644\u062d\u0641\u0638 \u0639\u0646 \u0638\u0647\u0631 \u0642\u0644\u0628. \u0643\u0645\u0627 \u064a\u0645\u0643\u0646 \u0627\u062e\u062a\u064a\u0627\u0631 \u0634\u0631\u0648\u0637 \u0627\u0644\u062a\u0642\u064a\u064a\u0645 \u0628\u0634\u0643\u0644\u064d \u0645\u064f\u0641\u0636\u064e\u0651\u0644: \u0645\u062b\u0644 \u0627\u0633\u062a\u062e\u062f\u0627\u0645 \u0645\u062c\u0645\u0648\u0639\u0629 \u0641\u0631\u0639\u064a\u0629 \u0645\u062e\u062a\u0644\u0641\u0629\u060c \u0623\u0648 \u0639\u062f\u062f \u0623\u0643\u0628\u0631 \u0645\u0646 \u0627\u0644\u0645\u062d\u0627\u0648\u0644\u0627\u062a\u060c \u0623\u0648 \u0627\u0644\u0627\u0633\u062a\u0641\u0627\u062f\u0629 \u0645\u0646 \u0623\u062f\u0648\u0627\u062a \u0645\u0639\u064a\u0646\u0629 \u0639\u0646\u062f \u0627\u0644\u0645\u0642\u0627\u0631\u0646\u0629 \u0628\u0646\u0645\u0648\u0630\u062c \u0644\u0627 \u064a\u0633\u062a\u062e\u062f\u0645\u0647\u0627. \u0648\u062a\u062e\u0641\u0650\u0651\u0641 \u0644\u0648\u0627\u0626\u062d \u0627\u0644\u062a\u0635\u0646\u064a\u0641 \u0627\u0644\u0645\u0633\u062a\u0642\u0644\u0629 \u0648\u0627\u0644\u0645\u0624\u0634\u0631\u0627\u062a \u0627\u0644\u0645\u0631\u0643\u0628\u0629 \u0645\u0646 \u0643\u0644\u0627 \u0627\u0644\u0645\u0634\u0643\u0644\u062a\u064a\u0646.
\n\u0648\u062a\u062e\u0641\u0650\u0651\u0641 \u0644\u0648\u0627\u0626\u062d \u0627\u0644\u062a\u0635\u0646\u064a\u0641 \u0627\u0644\u0645\u0633\u062a\u0642\u0644\u0629 \u0648\u0627\u0644\u0645\u0624\u0634\u0631\u0627\u062a \u0627\u0644\u0645\u0631\u0643\u0628\u0629 \u0645\u0646 \u0643\u0644\u0627 \u0627\u0644\u0645\u0634\u0643\u0644\u062a\u064a\u0646.
\n\u0648\u062a\u062e\u0641\u0650\u0651\u0641 \u0644\u0648\u0627\u0626\u062d \u0627\u0644\u062a\u0635\u0646\u064a\u0641 \u0627\u0644\u0645\u0633\u062a\u0642\u0644\u0629 \u0648\u0627\u0644\u0645\u0624\u0634\u0631\u0627\u062a \u0627\u0644\u0645\u0631\u0643\u0628\u0629 \u0645\u0646 \u0643\u0644\u0627 \u0627\u0644\u0645\u0634\u0643\u0644\u062a\u064a\u0646.<\/p>\n<\/details>\n

\n\u0647\u0644 \u062a\u062a\u0646\u0628\u064e\u0651\u0623 \u0646\u062a\u0627\u0626\u062c \u0627\u0644\u0645\u0642\u0627\u064a\u064a\u0633 \u0628\u0623\u062f\u0627\u0621 \u0627\u0644\u0646\u0645\u0648\u0630\u062c \u0641\u064a \u0627\u0644\u0639\u0627\u0644\u0645 \u0627\u0644\u062d\u0642\u064a\u0642\u064a\u061f<\/summary>\n

\u062c\u0632\u0626\u064a\u064b\u0651\u0627. \u0641\u0627\u0644\u0645\u0642\u0627\u064a\u064a\u0633 \u0627\u0644\u0645\u064f\u0635\u0645\u064e\u0651\u0645\u0629 \u0644\u062a\u0643\u0648\u0646 \u0645\u064f\u0637\u0627\u0628\u0642\u0629\u064b \u0644\u0644\u0645\u0647\u0627\u0645 \u0627\u0644\u0648\u0627\u0642\u0639\u064a\u0629\u060c \u0645\u062b\u0644 \u00abSWE-bench Verified\u00bb \u0648\u00abOSWorld\u00bb\u060c \u062a\u062a\u0646\u0628\u064e\u0651\u0623 \u0628\u062f\u0642\u0629 \u0645\u0639\u0642\u0648\u0644\u0629 \u0644\u0623\u0646 \u0627\u0644\u0645\u0647\u0645\u0629 \u0627\u0644\u0645\u064f\u062e\u062a\u0628\u0631\u0629 \u0647\u064a \u0628\u0627\u0644\u0641\u0639\u0644 \u0627\u0644\u0645\u0647\u0645\u0629 \u0627\u0644\u0648\u0627\u0642\u0639\u064a\u0629 \u0646\u0641\u0633\u0647\u0627.
\n\u0623\u0645\u0627 \u0645\u0642\u0627\u064a\u064a\u0633 \u0627\u0644\u0627\u062e\u062a\u064a\u0627\u0631\u0627\u062a \u0627\u0644\u0645\u062a\u0639\u062f\u062f\u0629 \u0627\u0644\u0623\u0643\u0627\u062f\u064a\u0645\u064a\u0629 \u0641\u0647\u064a \u062a\u062a\u0646\u0628\u064e\u0651\u0623 \u0628\u0634\u0643\u0644 \u0636\u0639\u064a\u0641\u060c \u0644\u0623\u0646 \u0627\u0644\u0625\u062c\u0627\u0628\u0629 \u0639\u0646 \u0623\u0633\u0626\u0644\u0629 \u0627\u0644\u0627\u0645\u062a\u062d\u0627\u0646\u0627\u062a \u0644\u064a\u0633\u062a \u0627\u0644\u0645\u0647\u0645\u0629 \u0627\u0644\u062a\u064a \u062a\u0624\u062f\u064a\u0647\u0627 \u0645\u0639\u0638\u0645 \u0627\u0644\u062a\u0637\u0628\u064a\u0642\u0627\u062a \u0639\u0627\u062f\u0629\u064b.
\n\u0648\u0623\u0641\u0636\u0644 \u0645\u0624\u0634\u0651\u0631 \u062a\u0646\u0628\u064f\u0651\u0626\u064a \u0644\u0627 \u064a\u0632\u0627\u0644 \u0647\u0648 \u0627\u0644\u062a\u0642\u064a\u064a\u0645 \u0627\u0644\u0630\u064a \u062a\u0642\u0648\u0645 \u0623\u0646\u062a \u0628\u0628\u0646\u0627\u0626\u0647 \u0639\u0644\u0649 \u0628\u064a\u0627\u0646\u0627\u062a\u0643 \u0627\u0644\u062e\u0627\u0635\u0629.
\n\u0648\u0623\u0641\u0636\u0644 \u0645\u0624\u0634\u0651\u0631 \u062a\u0646\u0628\u064f\u0651\u0626\u064a \u0644\u0627 \u064a\u0632\u0627\u0644 \u0647\u0648 \u0627\u0644\u062a\u0642\u064a\u064a\u0645 \u0627\u0644\u0630\u064a \u062a\u0642\u0648\u0645 \u0623\u0646\u062a \u0628\u0628\u0646\u0627\u0626\u0647 \u0639\u0644\u0649 \u0628\u064a\u0627\u0646\u0627\u062a\u0643 \u0627\u0644\u062e\u0627\u0635\u0629.<\/p>\n<\/details>\n

\n\u0623\u064a \u0646\u0645\u0648\u0630\u062c \u0645\u0641\u062a\u0648\u062d \u0627\u0644\u0648\u0632\u0646 \u064a\u062d\u0642\u0642 \u0623\u0639\u0644\u0649 \u062f\u0631\u062c\u0629\u061f<\/summary>\n

\u0646\u0645\u0648\u0630\u062c \u0643\u064a\u0645\u064a K3 \u0628\u0646\u062a\u064a\u062c\u0629 \u0665\u0667 \u0639\u0644\u0649 \u0645\u0624\u0634\u0631 \u0627\u0644\u0630\u0643\u0627\u0621 \u0627\u0644\u0627\u0635\u0637\u0646\u0627\u0639\u064a \u0645\u0646 Artificial Analysis\u060c \u0648\u0647\u064a \u0623\u0641\u0636\u0644 \u0646\u062a\u064a\u062c\u0629 \u0645\u062d\u0642\u064e\u0651\u0642\u0629 \u0644\u0646\u0645\u0648\u0630\u062c \u0645\u0641\u062a\u0648\u062d \u0627\u0644\u0648\u0632\u0646 \u062d\u062a\u0649 \u0627\u0644\u0622\u0646\u060c \u0648\u062a\u0642\u0644\u064f\u0651 \u0628\u0640\u0664 \u0646\u0642\u0627\u0637 \u0641\u0642\u0637 \u0639\u0646 \u0623\u0641\u0636\u0644 \u0646\u0645\u0648\u0630\u062c \u0645\u063a\u0644\u0642.
\n\u0627\u0644\u062c\u062f\u0648\u0644 \u0627\u0644\u0642\u064a\u0627\u062f\u064a \u0623\u0639\u0644\u0627\u0647 \u064a\u064f\u062d\u062f\u0650\u0651\u062f \u0643\u0644 \u0625\u062f\u062e\u0627\u0644 \u0645\u0641\u062a\u0648\u062d \u0627\u0644\u0648\u0632\u0646\u060c \u0648\u0627\u0644\u0645\u062c\u0645\u0648\u0639\u0629 \u0627\u0644\u0643\u0627\u0645\u0644\u0629 \u0642\u0627\u0628\u0644\u0629 \u0644\u0644\u0641\u0631\u0632 \u0641\u064a
\n\u0627\u0644\u062c\u062f\u0648\u0644 \u0623\u062f\u0646\u0627\u0647
\n\u0644\u0648\u062d\u0629 \u062a\u0635\u0646\u064a\u0641 \u0627\u0644\u0646\u0645\u0627\u0630\u062c \u0627\u0644\u0644\u063a\u0648\u064a\u0629 \u0627\u0644\u0643\u0628\u064a\u0631\u0629 (LLM)<\/a>.<\/p>\n<\/details>\n<\/div>","protected":false},"excerpt":{"rendered":"

An AI benchmark is a fixed set of tasks used to score a model’s ability, so that different models can […]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"class_list":["post-2102","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/convly.ai\/ar\/wp-json\/wp\/v2\/pages\/2102","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/convly.ai\/ar\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/convly.ai\/ar\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/convly.ai\/ar\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/convly.ai\/ar\/wp-json\/wp\/v2\/comments?post=2102"}],"version-history":[{"count":0,"href":"https:\/\/convly.ai\/ar\/wp-json\/wp\/v2\/pages\/2102\/revisions"}],"wp:attachment":[{"href":"https:\/\/convly.ai\/ar\/wp-json\/wp\/v2\/media?parent=2102"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}