98.1 Ukrainian (Ukraine) ukrainian_polywrite
ukrainian open generation
open_quality_score: 98.1sample_len: 154.0
open_quality_score: 98.1sample_len: 154.0
96.5 Albanian (Albania) albanian_polywrite
albanian open generation
open_quality_score: 96.5sample_len: 155.0
open_quality_score: 96.5sample_len: 155.0
91.8 Ukrainian (Ukraine) ukrainian_belebele
ukrainian mcq
f1_macro: 91.8sample_len: 900.0
f1_macro: 91.8sample_len: 900.0
91.8 Portuguese (Portugal) portuguese_enem
portuguese mcq
exact_match: 91.8sample_len: 1432.0
exact_match: 91.8sample_len: 1432.0
90.5 French (France) french_sib200
french classification
f1_macro: 90.5sample_len: 204.0
f1_macro: 90.5sample_len: 204.0
90.1 Portuguese (Portugal) portuguese_bluex
portuguese mcq
exact_match: 90.1sample_len: 724.0
exact_match: 90.1sample_len: 724.0
89.6 Albanian (Albania) albanian_belebele
albanian mcq
f1_macro: 89.6sample_len: 900.0
f1_macro: 89.6sample_len: 900.0
89.3 Swahili (Tanzania) swahili_sib200
swahili classification
f1_macro: 89.3sample_len: 204.0
f1_macro: 89.3sample_len: 204.0
88.9 Ukrainian (Ukraine) ukrainian_sib200
ukrainian classification
f1_macro: 88.9sample_len: 204.0
f1_macro: 88.9sample_len: 204.0
88.8 English (US) english_mgsm
english math
exact_match: 88.8sample_len: 250.0
exact_match: 88.8sample_len: 250.0
87.2 Arabic (Saudi Arabia) arabic_sib200
arabic classification
f1_macro: 87.2sample_len: 204.0
f1_macro: 87.2sample_len: 204.0
87.2 Albanian (Albania) albanian_sib200
albanian classification
f1_macro: 87.2sample_len: 204.0
f1_macro: 87.2sample_len: 204.0
86.8 English (US) english_gsm8k
english math
exact_match: 86.8sample_len: 1319.0
exact_match: 86.8sample_len: 1319.0
86.0 Hausa (Nigeria) hausa_sib200
hausa classification
f1_macro: 86.0sample_len: 204.0
f1_macro: 86.0sample_len: 204.0
85.4 Portuguese (Portugal) portuguese_hatebr
portuguese classification
f1_macro: 85.4sample_len: 1400.0
f1_macro: 85.4sample_len: 1400.0
84.7 Ukrainian (Ukraine) ukrainian_global_mmlu
ukrainian mcq
f1_macro: 84.7sample_len: 2850.0
f1_macro: 84.7sample_len: 2850.0
82.8 Albanian (Albania) albanian_global_mmlu
albanian mcq
f1_macro: 82.8sample_len: 400.0
f1_macro: 82.8sample_len: 400.0
82.7 Urdu (Pakistan) urdu_uquad
urdu qa
llm_judge_score: 82.7sample_len: 139.0
llm_judge_score: 82.7sample_len: 139.0
81.8 Igbo (Nigeria) igbo_sib200
igbo classification
f1_macro: 81.8sample_len: 204.0
f1_macro: 81.8sample_len: 204.0
78.6 Yoruba (Nigeria) yoruba_sib200
yoruba classification
f1_macro: 78.6sample_len: 204.0
f1_macro: 78.6sample_len: 204.0
74.0 Albanian (Albania) albanian_aya
albanian open generation
llm_judge_score: 74.0sample_len: 200.0
llm_judge_score: 74.0sample_len: 200.0
74.0 English (US) english_mmlu_pro
english mmlu pro
exact_match: 74.0sample_len: 2100.0
exact_match: 74.0sample_len: 2100.0
71.2 French (France) french_mgsm
french math
exact_match: 71.2sample_len: 250.0
exact_match: 71.2sample_len: 250.0
70.2 Portuguese (Portugal) portuguese_tweetsentbr
portuguese classification
f1_macro: 70.2sample_len: 2010.0
f1_macro: 70.2sample_len: 2010.0
70.0 Ukrainian (Ukraine) ukrainian_zno
ukrainian mcq
f1_macro: 70.0sample_len: 751.0
f1_macro: 70.0sample_len: 751.0
69.6 Swahili (Tanzania) swahili_afrimgsm
swahili afrimgsm
exact_match: 69.6sample_len: 250.0
exact_match: 69.6sample_len: 250.0
68.9 Yoruba (Nigeria) yoruba_naijasenti
yoruba sentiment
f1_macro: 68.9sample_len: 4515.0
f1_macro: 68.9sample_len: 4515.0
68.6 Hausa (Nigeria) hausa_afrixnli
hausa nli
f1_macro: 68.6sample_len: 600.0
f1_macro: 68.6sample_len: 600.0
67.8 Hausa (Nigeria) hausa_naijasenti
hausa sentiment
f1_macro: 67.8sample_len: 5303.0
f1_macro: 67.8sample_len: 5303.0
66.3 Urdu (Pakistan) urdu_freshqa
urdu qa
llm_judge_score: 66.3sample_len: 323.0
llm_judge_score: 66.3sample_len: 323.0
65.8 Portuguese (Portugal) portuguese_hate_speech
portuguese classification
f1_macro: 65.8sample_len: 851.0
f1_macro: 65.8sample_len: 851.0
65.6 Hausa (Nigeria) hausa_afrimgsm
hausa afrimgsm
exact_match: 65.6sample_len: 250.0
exact_match: 65.6sample_len: 250.0
65.0 Swahili (Tanzania) swahili_afrixnli
swahili nli
f1_macro: 65.0sample_len: 600.0
f1_macro: 65.0sample_len: 600.0
64.9 Igbo (Nigeria) igbo_afrixnli
igbo nli
f1_macro: 64.9sample_len: 600.0
f1_macro: 64.9sample_len: 600.0
64.9 Spanish (Spain) spanish_xquad_es
spanish xquad es
exact_match: 54.4f1: 75.4sample_len: 1190.0
exact_match: 54.4f1: 75.4sample_len: 1190.0
62.9 Portuguese (Portugal) portuguese_oab_exams
portuguese mcq
exact_match: 62.9sample_len: 2210.0
exact_match: 62.9sample_len: 2210.0
62.7 Igbo (Nigeria) igbo_naijasenti
igbo sentiment
f1_macro: 62.7sample_len: 3682.0
f1_macro: 62.7sample_len: 3682.0
60.8 Yoruba (Nigeria) yoruba_afrimgsm
yoruba afrimgsm
exact_match: 60.8sample_len: 250.0
exact_match: 60.8sample_len: 250.0
60.5 Yoruba (Nigeria) yoruba_afrixnli
yoruba nli
f1_macro: 60.5sample_len: 600.0
f1_macro: 60.5sample_len: 600.0
58.4 Ukrainian (Ukraine) ukrainian_squad
ukrainian qa
exact_match: 46.2f1: 70.6sample_len: 3812.0
exact_match: 46.2f1: 70.6sample_len: 3812.0
57.8 French (France) french_fquad
french qa
exact_match: 47.3f1: 68.3sample_len: 400.0
exact_match: 47.3f1: 68.3sample_len: 400.0
53.2 Igbo (Nigeria) igbo_afrimgsm
igbo afrimgsm
exact_match: 53.2sample_len: 250.0
exact_match: 53.2sample_len: 250.0
50.1 Urdu (Pakistan) urdu_facttool_qa
urdu claim
f1_macro: 50.1sample_len: 160.0
f1_macro: 50.1sample_len: 160.0
44.6 Urdu (Pakistan) urdu_factcheckbench
urdu claim
f1_macro: 44.6sample_len: 387.0
f1_macro: 44.6sample_len: 387.0
44.3 English (US) ifeval
ifeval
inst_level_loose_acc: 56.0inst_level_strict_acc: 45.8prompt_level_loose_acc: 44.0prompt_level_strict_acc: 31.6sample_len: 541.0
inst_level_loose_acc: 56.0inst_level_strict_acc: 45.8prompt_level_loose_acc: 44.0prompt_level_strict_acc: 31.6sample_len: 541.0
44.3 Urdu (Pakistan) urdu_bingcheck
urdu claim
f1_macro: 44.3sample_len: 102.0
f1_macro: 44.3sample_len: 102.0
40.9 Arabic (Saudi Arabia) arabic_tydiqa
arabic qa
exact_match: 30.0f1: 51.7sample_len: 921.0
exact_match: 30.0f1: 51.7sample_len: 921.0
40.8 English (US) english_belebele
english mcq
f1_macro: 40.8sample_len: 900.0
f1_macro: 40.8sample_len: 900.0
38.1 French (France) french_belebele
french mcq
f1_macro: 38.1sample_len: 900.0
f1_macro: 38.1sample_len: 900.0
37.0 Arabic (Saudi Arabia) arabic_belebele
arabic mcq
f1_macro: 37.0sample_len: 900.0
f1_macro: 37.0sample_len: 900.0
33.7 Spanish (Spain) spanish_belebele
spanish mcq
f1_macro: 33.7sample_len: 900.0
f1_macro: 33.7sample_len: 900.0
33.6 Urdu (Pakistan) urdu_fake_news
urdu classification
f1_macro: 33.6sample_len: 300.0
f1_macro: 33.6sample_len: 300.0
33.5 Urdu (Pakistan) urdu_emotion_class
urdu classification
f1_macro: 33.5sample_len: 200.0
f1_macro: 33.5sample_len: 200.0
31.2 Spanish (Spain) spanish_global_mmlu
spanish mcq
f1_macro: 31.2sample_len: 400.0
f1_macro: 31.2sample_len: 400.0
30.3 French (France) french_mmmlu
french mcq
f1_macro: 30.3sample_len: 14042.0
f1_macro: 30.3sample_len: 14042.0
29.2 Igbo (Nigeria) igbo_afriqa
igbo qa
exact_match: 24.7f1: 33.7sample_len: 409.0
exact_match: 24.7f1: 33.7sample_len: 409.0
25.2 Hausa (Nigeria) hausa_afrimmlu
hausa mcq
f1_macro: 25.2sample_len: 500.0
f1_macro: 25.2sample_len: 500.0
23.5 Yoruba (Nigeria) yoruba_afrimmlu
yoruba mcq
f1_macro: 23.5sample_len: 500.0
f1_macro: 23.5sample_len: 500.0
22.8 Hausa (Nigeria) hausa_belebele
hausa mcq
f1_macro: 22.8sample_len: 900.0
f1_macro: 22.8sample_len: 900.0
22.3 Igbo (Nigeria) igbo_afrimmlu
igbo mcq
f1_macro: 22.3sample_len: 500.0
f1_macro: 22.3sample_len: 500.0
22.1 Igbo (Nigeria) igbo_belebele
igbo mcq
f1_macro: 22.1sample_len: 900.0
f1_macro: 22.1sample_len: 900.0
19.2 Hausa (Nigeria) hausa_afriqa
hausa qa
exact_match: 17.3f1: 21.0sample_len: 300.0
exact_match: 17.3f1: 21.0sample_len: 300.0
19.1 Yoruba (Nigeria) yoruba_belebele
yoruba mcq
f1_macro: 19.1sample_len: 900.0
f1_macro: 19.1sample_len: 900.0
17.5 Urdu (Pakistan) urdu_simpleqa
urdu qa
llm_judge_score: 17.5sample_len: 200.0
llm_judge_score: 17.5sample_len: 200.0
17.1 Arabic (Saudi Arabia) arabic_aratrust
arabic mcq
f1_macro: 17.1sample_len: 522.0
f1_macro: 17.1sample_len: 522.0
15.7 Yoruba (Nigeria) yoruba_afriqa
yoruba qa
exact_match: 12.7f1: 18.8sample_len: 332.0
exact_match: 12.7f1: 18.8sample_len: 332.0
14.7 Arabic (Saudi Arabia) arabic_mmlu
arabic mcq
f1_macro: 14.7sample_len: 14316.0
f1_macro: 14.7sample_len: 14316.0