Skip to content

Latest commit

 

History

History
1874 lines (1868 loc) · 116 KB

faroese-nlu.md

File metadata and controls

1874 lines (1868 loc) · 116 KB
layout title
leaderboard
Faroese NLU 🇫🇴
Last updated: 04/11/2024 17:09:24 CET
Include merged models
Model ID Parameters Vocabulary size Context Commercial Speed Rank FoNE ScaLA-fo FoQA FoNE version ScaLA-fo version FoQA version
vesteinn/FoBERT 124 50 512 True 15,623 ± 2,828 / 3,737 ± 1,191 1.58 91.31 ± 0.69 / 91.79 ± 0.47 64.39 ± 1.55 / 81.38 ± 1.04 27.09 ± 1.55 / 36.68 ± 1.88 0.0.0 0.0.0 13.0.0
gpt-4-1106-preview (few-shot, val) unknown 100 128000 True 576 ± 221 / 81 ± 28 2.06 86.51 ± 2.01 / 85.01 ± 2.45 35.09 ± 5.70 / 65.72 ± 3.27 58.65 ± 3.25 / 81.93 ± 2.20 12.5.1 12.5.1 13.0.0
NbAiLab/nb-roberta-base-scandi-1e4 277 250 512 True 15,074 ± 2,990 / 3,347 ± 1,080 2.23 90.52 ± 0.47 / 90.83 ± 0.40 44.99 ± 11.76 / 71.54 ± 5.83 25.14 ± 1.72 / 34.64 ± 1.98 0.0.0 0.0.0 13.0.0
microsoft/mdeberta-v3-base 278 251 512 True 20,637 ± 3,925 / 4,497 ± 1,502 2.33 88.60 ± 0.60 / 89.37 ± 0.54 46.81 ± 2.12 / 72.76 ± 1.40 20.96 ± 1.19 / 30.59 ± 1.61 0.0.0 0.0.0 13.0.0
sentence-transformers/use-cmlm-multilingual 470 501 512 True 30,231 ± 8,171 / 4,863 ± 1,598 2.60 88.81 ± 0.65 / 89.12 ± 0.56 30.92 ± 8.65 / 63.05 ± 4.66 25.48 ± 2.79 / 37.21 ± 3.83 0.0.0 0.0.0 13.0.0
pere/roberta-base-exp-32 277 250 512 True 15,081 ± 2,950 / 3,365 ± 1,092 2.61 90.60 ± 0.45 / 90.78 ± 0.45 22.86 ± 13.14 / 59.92 ± 6.74 24.90 ± 1.52 / 33.83 ± 1.68 0.0.0 0.0.0 13.0.0
vesteinn/ScandiBERT-no-faroese 124 50 512 True 15,436 ± 2,820 / 3,704 ± 1,187 2.73 88.14 ± 0.58 / 88.89 ± 0.52 27.71 ± 9.67 / 61.60 ± 5.69 20.47 ± 2.44 / 29.37 ± 3.28 0.0.0 0.0.0 13.0.0
gpt-4o-mini-2024-07-18 (few-shot) unknown 200 128126 True 1,171 ± 378 / 120 ± 39 2.76 77.59 ± 2.14 / 63.75 ± 5.57 14.58 ± 2.56 / 51.54 ± 2.48 51.20 ± 1.49 / 75.23 ± 1.04 12.11.0 12.11.0 13.0.0
google/rembert 575 250 256 True 11,736 ± 2,822 / 2,102 ± 677 2.87 87.35 ± 0.94 / 87.65 ± 0.94 14.65 ± 11.12 / 46.36 ± 10.23 36.10 ± 1.91 / 48.41 ± 2.09 0.0.0 0.0.0 13.0.0
setu4993/LaBSE 470 501 512 True 25,418 ± 6,435 / 4,536 ± 1,452 2.95 89.16 ± 0.66 / 89.62 ± 0.62 22.76 ± 10.57 / 60.04 ± 5.76 30.55 ± 1.05 / 41.85 ± 1.13 0.0.0 0.0.0 13.0.0
ltg/norbert3-base 124 50 512 True 11,405 ± 1,970 / 2,856 ± 917 3.10 86.94 ± 0.47 / 87.58 ± 0.46 12.35 ± 8.96 / 53.00 ± 5.17 24.57 ± 2.18 / 34.06 ± 2.82 12.7.0 12.7.0 13.0.0
jonfd/electra-small-nordic 22 96 128 True 5,989 ± 120 / 3,809 ± 1,230 3.17 85.80 ± 0.25 / 86.58 ± 0.26 30.88 ± 2.36 / 63.79 ± 1.33 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
Geotrend/bert-base-25lang-cased 151 85 512 True 13,908 ± 3,201 / 2,700 ± 872 3.27 86.09 ± 1.03 / 86.85 ± 1.02 15.24 ± 6.84 / 50.54 ± 4.85 14.82 ± 1.50 / 22.99 ± 1.63 0.0.0 0.0.0 13.0.0
gpt-4o-2024-05-13 (few-shot, val) unknown 200 128000 True 916 ± 329 / 114 ± 38 3.32 81.86 ± 2.08 / 67.98 ± 3.64 -0.97 ± 2.82 / 33.64 ± 0.85 56.45 ± 2.88 / 78.76 ± 1.86 12.10.0 12.10.0 13.0.0
mideind/IceBERT-xlmr-ic3 277 250 512 True 11,004 ± 2,244 / 2,324 ± 761 3.34 87.79 ± 0.40 / 88.46 ± 0.31 22.51 ± 10.65 / 55.58 ± 8.05 11.16 ± 3.59 / 16.82 ± 5.50 0.0.0 0.0.0 13.0.0
meta-llama/Meta-Llama-3-8B-Instruct (few-shot) 8030 128 8192 True 4,909 ± 1,215 / 978 ± 319 3.50 73.76 ± 1.34 / 70.21 ± 2.63 3.47 ± 1.97 / 49.04 ± 1.89 49.53 ± 2.28 / 73.08 ± 1.05 12.6.1 12.6.1 13.0.0
mideind/IceBERT-large 406 50 512 True 5,677 ± 719 / 1,886 ± 604 3.53 86.84 ± 0.77 / 87.51 ± 0.72 9.82 ± 3.95 / 50.49 ± 3.45 21.19 ± 2.49 / 30.11 ± 2.76 0.0.0 0.0.0 13.0.0
AI-Sweden-Models/roberta-large-1160k 354 50 512 True 14,014 ± 2,384 / 3,625 ± 1,146 3.54 88.24 ± 0.58 / 88.48 ± 0.54 1.73 ± 1.70 / 42.32 ± 5.16 35.08 ± 2.58 / 47.03 ± 3.42 10.0.1 10.0.1 13.0.0
intfloat/multilingual-e5-large 559 250 512 True 6,732 ± 1,273 / 1,633 ± 523 3.57 88.39 ± 0.86 / 88.75 ± 0.75 2.85 ± 1.32 / 48.43 ± 2.29 31.03 ± 1.94 / 43.72 ± 2.89 0.0.0 0.0.0 13.0.0
Nexusflow/Starling-LM-7B-beta (few-shot) 7242 32 8192 False 5,876 ± 1,021 / 1,677 ± 546 3.58 66.98 ± 1.97 / 60.82 ± 2.81 5.80 ± 2.04 / 42.89 ± 2.71 44.88 ± 1.73 / 63.94 ± 1.35 12.5.2 12.5.2 13.0.0
mistralai/Ministral-8B-Instruct-2410 (few-shot) 8020 131 32768 True 5,821 ± 1,090 / 1,561 ± 506 3.60 67.52 ± 2.82 / 61.64 ± 3.39 3.35 ± 2.55 / 46.55 ± 3.36 49.62 ± 1.90 / 69.57 ± 1.48 13.0.0 13.0.0 13.0.0
intfloat/multilingual-e5-base 278 250 512 True 14,965 ± 2,890 / 3,322 ± 1,074 3.61 87.44 ± 0.35 / 88.01 ± 0.28 7.38 ± 6.28 / 47.69 ± 4.51 14.80 ± 3.56 / 22.12 ± 5.23 12.6.1 12.6.1 13.0.0
meta-llama/Llama-3.1-8B (few-shot) 8030 128 131200 True 2,439 ± 459 / 703 ± 219 3.64 65.43 ± 3.19 / 63.09 ± 4.04 1.12 ± 2.01 / 35.45 ± 2.25 51.86 ± 1.50 / 73.55 ± 1.19 12.11.0 12.11.0 13.0.0
meta-llama/Llama-3.1-8B-Instruct (few-shot) 8030 128 131200 True 2,411 ± 465 / 686 ± 215 3.67 71.04 ± 1.85 / 70.33 ± 2.15 1.59 ± 1.29 / 33.68 ± 0.38 46.33 ± 1.15 / 70.86 ± 1.12 13.0.0 13.0.0 13.0.0
KBLab/megatron-bert-large-swedish-cased-165k 369 64 512 True 7,138 ± 1,111 / 2,067 ± 660 3.70 82.76 ± 0.95 / 83.43 ± 0.88 7.58 ± 3.56 / 52.05 ± 1.14 16.13 ± 1.68 / 23.78 ± 2.02 0.0.0 0.0.0 13.0.0
FacebookAI/xlm-roberta-large 559 250 512 True 17,897 ± 3,921 / 3,463 ± 1,141 3.71 87.85 ± 0.95 / 88.21 ± 0.87 1.17 ± 2.11 / 40.94 ± 5.07 27.72 ± 1.48 / 39.88 ± 2.04 0.0.0 0.0.0 13.0.0
mhenrichsen/hestenettetLM (few-shot) 7242 32 32768 True 5,160 ± 804 / 1,654 ± 516 3.72 62.82 ± 3.50 / 58.05 ± 3.79 4.96 ± 2.36 / 43.40 ± 4.95 44.02 ± 1.87 / 60.50 ± 2.26 12.5.2 12.3.2 13.0.0
CohereForAI/aya-expanse-8b (few-shot) 8028 256 8192 False 5,581 ± 1,066 / 1,471 ± 483 3.74 68.63 ± 1.70 / 41.26 ± 2.21 1.95 ± 2.22 / 42.77 ± 3.69 45.04 ± 1.74 / 63.06 ± 1.68 13.0.0 13.0.0 13.0.0
intfloat/multilingual-e5-large-instruct 560 250 514 True 5,947 ± 1,301 / 1,129 ± 374 3.74 88.64 ± 0.34 / 89.11 ± 0.29 2.05 ± 2.30 / 47.88 ± 2.17 24.09 ± 4.66 / 37.60 ± 4.39 13.0.0 13.0.0 13.0.0
meta-llama/Meta-Llama-3-8B (few-shot) 8030 128 8192 True 4,687 ± 1,121 / 967 ± 313 3.74 61.11 ± 4.21 / 58.55 ± 4.19 2.02 ± 1.68 / 39.88 ± 3.56 50.34 ± 1.74 / 71.74 ± 1.27 12.6.1 12.6.1 13.0.0
NbAiLab/nb-sbert-base 177 120 512 True 17,757 ± 3,883 / 3,864 ± 1,237 3.80 86.20 ± 0.50 / 86.91 ± 0.46 11.80 ± 7.52 / 53.06 ± 5.04 7.47 ± 1.01 / 11.22 ± 1.50 12.10.5 12.10.5 13.0.0
google-bert/bert-base-multilingual-cased 177 120 512 True 14,083 ± 3,264 / 2,738 ± 889 3.80 86.36 ± 0.57 / 87.14 ± 0.52 6.08 ± 4.57 / 48.47 ± 3.93 16.26 ± 1.27 / 24.64 ± 1.39 0.0.0 0.0.0 13.0.0
mistralai/Mistral-7B-v0.1 (few-shot) 7242 32 32768 True 2,657 ± 524 / 880 ± 278 3.81 62.63 ± 3.44 / 57.85 ± 3.72 2.84 ± 1.84 / 42.62 ± 4.53 44.06 ± 1.93 / 60.40 ± 2.37 9.1.2 9.1.2 13.0.0
KBLab/megatron-bert-base-swedish-cased-125k 135 64 512 True 15,763 ± 2,523 / 4,238 ± 1,370 3.83 80.61 ± 0.91 / 81.31 ± 0.89 9.60 ± 3.82 / 52.30 ± 1.73 11.57 ± 0.86 / 17.37 ± 1.12 12.7.0 12.7.0 13.0.0
mideind/IceBERT 124 50 512 True 16,697 ± 2,113 / 5,432 ± 1,749 3.83 86.50 ± 0.96 / 87.11 ± 0.90 10.13 ± 8.55 / 47.31 ± 7.04 5.35 ± 3.50 / 8.28 ± 5.36 0.0.0 0.0.0 13.0.0
mistralai/Mistral-7B-v0.3 (few-shot) 7248 33 32768 True 4,120 ± 976 / 926 ± 306 3.84 61.32 ± 4.26 / 59.28 ± 4.43 1.30 ± 1.64 / 45.10 ± 3.27 44.98 ± 1.62 / 62.38 ± 1.68 12.10.4 12.10.4 13.0.0
AI-Nordics/bert-large-swedish-cased 334 31 512 True 7,199 ± 1,139 / 2,051 ± 651 3.86 83.22 ± 0.71 / 83.91 ± 0.66 6.78 ± 3.31 / 51.53 ± 2.36 15.94 ± 1.67 / 22.00 ± 2.12 12.7.0 12.7.0 13.0.0
NbAiLab/nb-roberta-base-scandinavian 125 50 512 True 14,051 ± 3,289 / 2,704 ± 897 3.89 86.10 ± 0.64 / 86.75 ± 0.57 6.28 ± 2.41 / 49.62 ± 2.08 9.89 ± 2.08 / 14.41 ± 3.08 0.0.0 0.0.0 13.0.0
Geotrend/bert-base-da-cased 103 23 512 True 15,432 ± 2,838 / 3,642 ± 1,189 3.91 86.62 ± 0.43 / 87.31 ± 0.44 3.64 ± 3.82 / 49.77 ± 2.26 13.84 ± 1.99 / 21.44 ± 2.62 0.0.0 0.0.0 13.0.0
mideind/IceBERT-mC4-is 163 50 512 True 12,308 ± 1,614 / 3,851 ± 1,254 3.91 88.44 ± 0.35 / 89.11 ± 0.38 11.83 ± 4.90 / 48.95 ± 6.81 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
Mabeck/Heidrun-Mistral-7B-chat (few-shot) 7242 32 32768 False 5,822 ± 1,283 / 1,336 ± 430 3.93 63.40 ± 2.78 / 59.96 ± 2.91 2.16 ± 2.25 / 39.38 ± 4.12 40.73 ± 1.32 / 57.50 ± 1.39 12.7.0 12.7.0 13.0.0
ibm-granite/granite-3.0-8b-base (few-shot) 8171 49 4224 True 5,116 ± 943 / 1,436 ± 472 3.93 61.47 ± 1.57 / 58.60 ± 2.31 1.44 ± 1.85 / 34.71 ± 1.84 41.54 ± 1.49 / 58.90 ± 1.24 13.0.0 13.0.0 13.0.0
meta-llama/Llama-3.2-3B-Instruct (few-shot) 3213 128 131200 False 10,424 ± 2,641 / 2,081 ± 666 3.93 58.24 ± 3.73 / 47.44 ± 4.83 1.77 ± 2.13 / 45.44 ± 4.00 45.13 ± 1.12 / 64.71 ± 0.74 13.0.0 13.0.0 13.0.0
KennethEnevoldsen/dfm-sentence-encoder-medium 124 50 512 True 14,998 ± 2,549 / 3,833 ± 1,223 3.95 84.92 ± 0.70 / 85.65 ± 0.67 2.96 ± 2.66 / 45.42 ± 4.03 14.00 ± 2.44 / 20.08 ± 3.54 0.0.0 0.0.0 13.0.0
mistralai/Mistral-7B-Instruct-v0.2 (few-shot) 7242 32 32768 False 634 ± 179 / 110 ± 35 3.97 61.28 ± 2.98 / 54.02 ± 3.55 1.68 ± 1.41 / 50.06 ± 1.22 39.00 ± 1.21 / 59.78 ± 1.14 9.2.0 9.3.1 13.0.0
ibm-granite/granite-3.0-8b-instruct (few-shot) 8171 49 4096 True 5,090 ± 937 / 1,423 ± 466 3.98 59.96 ± 2.42 / 58.61 ± 2.57 2.24 ± 1.78 / 41.32 ± 3.70 39.52 ± 1.83 / 57.16 ± 2.10 13.0.0 13.0.0 13.0.0
KBLab/bert-base-swedish-cased-new 135 64 512 True 15,933 ± 2,541 / 4,289 ± 1,376 4.00 84.02 ± 0.86 / 84.66 ± 0.82 5.65 ± 3.07 / 49.17 ± 3.05 7.14 ± 2.31 / 10.87 ± 3.52 12.7.0 12.7.0 13.0.0
ibm-granite/granite-8b-code-base-4k (few-shot) 8055 49 4096 True 2,313 ± 423 / 682 ± 210 4.00 66.82 ± 2.36 / 65.84 ± 2.35 -0.36 ± 1.67 / 37.57 ± 3.35 36.47 ± 0.96 / 51.89 ± 1.05 13.0.0 13.0.0 13.0.0
flax-community/nordic-roberta-wiki 124 50 512 True 16,227 ± 2,650 / 4,252 ± 1,393 4.01 82.64 ± 0.63 / 83.27 ± 0.65 8.03 ± 2.01 / 51.96 ± 1.87 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
occiglot/occiglot-7b-eu5-instruct (few-shot) 7242 32 32768 False 2,088 ± 352 / 706 ± 214 4.01 60.37 ± 3.60 / 59.38 ± 3.85 0.00 ± 0.00 / 33.26 ± 0.34 43.69 ± 2.03 / 60.55 ± 1.76 12.5.2 12.3.1 13.0.0
mideind/IceBERT-ic3 124 50 512 True 12,119 ± 1,576 / 3,812 ± 1,242 4.02 87.22 ± 0.68 / 87.92 ± 0.59 6.23 ± 6.96 / 48.55 ± 4.87 2.98 ± 1.75 / 4.35 ± 2.54 0.0.0 0.0.0 13.0.0
KBLab/megatron-bert-base-swedish-cased-600k 135 64 512 True 15,726 ± 2,508 / 4,234 ± 1,365 4.03 81.02 ± 0.75 / 81.74 ± 0.75 4.00 ± 2.04 / 50.51 ± 1.11 12.98 ± 1.18 / 20.39 ± 1.75 12.7.0 12.7.0 13.0.0
vesteinn/XLMR-ENIS 125 50 512 True 10,711 ± 2,333 / 2,141 ± 689 4.03 87.09 ± 0.76 / 87.71 ± 0.73 3.09 ± 1.98 / 39.41 ± 4.45 5.75 ± 2.66 / 8.73 ± 4.07 0.0.0 0.0.0 13.0.0
ZurichNLP/unsup-simcse-xlm-roberta-base 277 250 512 True 34,520 ± 7,443 / 6,730 ± 2,224 4.04 84.14 ± 0.67 / 84.82 ± 0.64 1.33 ± 2.18 / 48.85 ± 1.23 14.20 ± 1.78 / 22.95 ± 2.71 12.6.1 12.6.1 13.0.0
CohereForAI/aya-23-8B (few-shot) 8028 256 8192 False 5,608 ± 1,062 / 1,472 ± 479 4.05 62.22 ± 2.10 / 58.79 ± 2.20 0.01 ± 2.43 / 37.74 ± 4.20 38.70 ± 1.22 / 54.70 ± 0.93 13.0.0 13.0.0 13.0.0
mideind/IceBERT-igc 124 50 512 True 12,551 ± 1,656 / 3,918 ± 1,274 4.05 83.82 ± 0.98 / 84.34 ± 0.88 4.93 ± 3.16 / 45.85 ± 3.66 5.17 ± 2.95 / 7.88 ± 4.65 0.0.0 0.0.0 13.0.0
AI-Sweden-Models/tyr (few-shot, val) 7242 32 32768 False 6,079 ± 1,051 / 1,760 ± 570 4.06 60.21 ± 3.50 / 58.44 ± 3.67 0.00 ± 0.00 / 33.10 ± 0.80 40.88 ± 2.10 / 59.65 ± 1.92 12.3.2 12.3.2 13.0.0
occiglot/occiglot-7b-eu5 (few-shot) 7242 32 32768 True 2,219 ± 427 / 717 ± 224 4.06 58.67 ± 3.95 / 58.47 ± 3.96 0.00 ± 0.00 / 33.26 ± 0.34 40.95 ± 1.80 / 55.82 ± 1.95 12.5.2 12.1.0 13.0.0
vesteinn/IceBERT 163 50 512 True 12,360 ± 1,611 / 3,858 ± 1,246 4.07 87.13 ± 0.58 / 87.70 ± 0.45 3.66 ± 3.33 / 40.81 ± 4.34 5.07 ± 3.59 / 8.05 ± 5.61 0.0.0 0.0.0 13.0.0
timpal0l/Mistral-7B-v0.1-flashback-v2 (few-shot) 7242 32 32768 True 5,054 ± 1,200 / 1,056 ± 339 4.10 58.96 ± 2.05 / 52.20 ± 2.50 0.00 ± 0.00 / 33.26 ± 0.34 39.20 ± 1.10 / 54.89 ± 1.25 12.5.3 12.5.3 13.0.0
KBLab/megatron-bert-large-swedish-cased-110k 369 64 512 True 7,075 ± 1,093 / 2,057 ± 661 4.11 82.36 ± 0.91 / 82.94 ± 0.90 5.20 ± 2.67 / 50.88 ± 1.55 2.65 ± 1.46 / 4.33 ± 2.24 0.0.0 0.0.0 13.0.0
google-bert/bert-base-multilingual-uncased 167 106 512 True 13,993 ± 3,217 / 2,752 ± 893 4.13 73.06 ± 1.51 / 72.61 ± 1.40 5.48 ± 2.42 / 49.18 ± 3.63 11.29 ± 2.19 / 18.32 ± 2.98 0.0.0 0.0.0 13.0.0
ibm-granite/granite-8b-code-instruct-4k (few-shot) 8055 49 4096 True 5,617 ± 995 / 1,623 ± 540 4.13 60.46 ± 2.53 / 58.29 ± 2.53 0.51 ± 0.78 / 36.20 ± 2.37 33.54 ± 1.00 / 49.02 ± 0.91 13.0.0 13.0.0 13.0.0
distilbert/distilbert-base-multilingual-cased 135 120 512 True 26,355 ± 5,946 / 5,266 ± 1,714 4.16 84.80 ± 0.85 / 85.49 ± 0.80 4.41 ± 4.92 / 50.84 ± 1.97 1.44 ± 1.03 / 2.60 ± 1.90 0.0.0 0.0.0 13.0.0
patrickvonplaten/norwegian-roberta-base 124 50 512 True 15,698 ± 2,699 / 3,891 ± 1,278 4.17 82.57 ± 0.93 / 83.35 ± 0.91 5.74 ± 3.53 / 48.75 ± 3.22 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
bineric/NorskGPT-Llama-7B-v0.1 (few-shot) 6738 32 4096 False 5,384 ± 879 / 1,746 ± 553 4.18 53.38 ± 3.96 / 53.04 ± 3.85 0.46 ± 2.01 / 44.83 ± 3.28 35.99 ± 1.01 / 51.25 ± 1.12 12.5.2 12.3.2 13.0.0
Twitter/twhin-bert-large 560 250 512 True 9,707 ± 1,664 / 2,549 ± 831 4.19 84.73 ± 1.49 / 85.19 ± 1.59 1.37 ± 2.46 / 39.78 ± 3.24 4.15 ± 4.39 / 5.74 ± 6.06 0.0.0 0.0.0 13.0.0
DeepPavlov/rubert-base-cased 177 120 512 True 15,785 ± 2,658 / 3,983 ± 1,289 4.20 83.15 ± 0.73 / 83.90 ± 0.74 3.21 ± 2.89 / 47.97 ± 3.56 1.53 ± 1.23 / 2.95 ± 2.23 0.0.0 0.0.0 13.0.0
Geotrend/distilbert-base-25lang-cased 109 85 512 True 26,099 ± 5,881 / 5,178 ± 1,665 4.20 83.21 ± 0.53 / 83.81 ± 0.45 2.37 ± 4.63 / 48.46 ± 3.19 1.35 ± 1.03 / 2.54 ± 1.88 12.6.1 12.6.1 13.0.0
vesteinn/DanskBERT 124 50 512 True 15,749 ± 2,665 / 4,014 ± 1,281 4.20 85.04 ± 0.57 / 85.72 ± 0.50 4.48 ± 1.63 / 44.45 ± 4.77 0.15 ± 0.29 / 0.21 ± 0.39 0.0.0 0.0.0 13.0.0
clips/mfaq 277 250 512 True 5,591 ± 187 / 3,349 ± 1,105 4.21 85.86 ± 0.51 / 86.55 ± 0.49 0.80 ± 2.32 / 42.37 ± 4.42 2.17 ± 2.62 / 2.93 ± 3.66 12.7.0 12.7.0 13.0.0
flax-community/swe-roberta-wiki-oscar 124 50 512 True 15,437 ± 2,628 / 3,834 ± 1,252 4.21 80.52 ± 0.76 / 81.35 ± 0.69 6.51 ± 3.60 / 51.81 ± 1.95 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
meta-llama/Llama-2-7b-chat-hf (few-shot) 6738 32 4096 False 2,643 ± 455 / 800 ± 247 4.21 59.77 ± 3.38 / 56.97 ± 4.30 -0.54 ± 1.61 / 36.94 ± 2.79 31.87 ± 2.20 / 46.21 ± 1.15 9.3.1 9.3.1 13.0.0
DDSC/roberta-base-danish 125 50 512 True 15,004 ± 2,964 / 3,290 ± 1,092 4.22 80.21 ± 0.95 / 80.80 ± 0.97 1.10 ± 2.44 / 48.16 ± 1.47 7.95 ± 1.36 / 11.84 ± 1.76 0.0.0 0.0.0 13.0.0
TrustLLMeu/baseline-7-8b_1t-tokens_llama (few-shot) 7800 100 4096 True 6,197 ± 1,118 / 1,730 ± 577 4.23 43.13 ± 3.96 / 42.16 ± 3.93 0.88 ± 1.93 / 34.15 ± 0.71 43.32 ± 1.32 / 63.19 ± 1.25 13.0.0 13.0.0 13.0.0
mistralai/Mistral-7B-Instruct-v0.1 (few-shot) 7242 32 32768 False 634 ± 179 / 110 ± 35 4.23 55.42 ± 2.12 / 46.41 ± 2.50 1.11 ± 2.41 / 36.79 ± 4.00 33.54 ± 1.29 / 50.80 ± 1.31 9.3.1 9.3.1 13.0.0
Geotrend/distilbert-base-en-no-cased 69 33 512 True 26,597 ± 6,036 / 5,271 ± 1,697 4.25 82.05 ± 0.46 / 82.67 ± 0.43 3.98 ± 2.94 / 50.53 ± 1.83 0.97 ± 1.01 / 1.92 ± 1.82 12.7.0 12.7.0 13.0.0
KB/bert-base-swedish-cased 124 50 512 True 16,181 ± 2,451 / 4,620 ± 1,507 4.25 82.76 ± 1.26 / 83.50 ± 1.20 3.98 ± 2.70 / 47.46 ± 2.35 0.00 ± 0.00 / 0.01 ± 0.02 0.0.0 0.0.0 13.0.0
sentence-transformers/stsb-xlm-r-multilingual 277 250 512 True 15,040 ± 2,953 / 3,417 ± 1,100 4.25 82.97 ± 0.83 / 83.62 ± 0.81 2.93 ± 1.96 / 46.51 ± 4.19 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
microsoft/infoxlm-base 277 250 512 True 34,735 ± 7,558 / 6,846 ± 2,312 4.27 85.58 ± 1.04 / 86.23 ± 1.03 0.35 ± 2.36 / 43.55 ± 4.58 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 118 250 512 True 29,201 ± 6,282 / 6,045 ± 2,027 4.27 82.24 ± 0.85 / 82.84 ± 0.78 2.84 ± 1.41 / 50.47 ± 1.13 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
dbmdz/bert-base-historic-multilingual-cased 110 32 512 True 20,047 ± 4,407 / 3,844 ± 1,259 4.28 80.45 ± 1.48 / 81.32 ± 1.42 2.52 ± 1.88 / 47.78 ± 1.67 0.58 ± 0.81 / 1.12 ± 1.62 0.0.0 0.0.0 13.0.0
Twitter/twhin-bert-base 279 250 512 True 11,514 ± 2,041 / 2,862 ± 918 4.29 84.00 ± 1.30 / 84.83 ± 1.28 1.94 ± 2.58 / 46.27 ± 2.74 0.02 ± 0.05 / 0.02 ± 0.05 12.6.1 12.6.1 13.0.0
MaLA-LM/emma-500-llama2-7b (few-shot) 6738 32 4096 True 6,275 ± 1,193 / 1,755 ± 578 4.31 40.18 ± 3.60 / 39.95 ± 3.38 0.31 ± 1.71 / 41.35 ± 4.05 41.60 ± 2.49 / 60.78 ± 2.12 13.0.0 13.0.0 13.0.0
cardiffnlp/twitter-xlm-roberta-base 277 250 512 True 34,475 ± 7,465 / 6,712 ± 2,223 4.31 83.96 ± 0.80 / 84.63 ± 0.76 1.05 ± 1.66 / 41.31 ± 4.91 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
deepset/gbert-base 109 31 512 True 37,268 ± 6,577 / 8,719 ± 2,865 4.31 80.48 ± 0.91 / 81.17 ± 0.85 0.60 ± 2.08 / 45.78 ± 2.54 2.63 ± 1.57 / 4.63 ± 2.53 0.0.0 0.0.0 13.0.0
microsoft/xlm-align-base 277 250 512 True 14,744 ± 2,870 / 3,265 ± 1,053 4.32 85.97 ± 1.12 / 86.52 ± 1.08 0.02 ± 1.38 / 44.65 ± 2.65 0.72 ± 0.93 / 1.00 ± 1.31 0.0.0 0.0.0 13.0.0
sentence-transformers/distilbert-multilingual-nli-stsb-quora-ranking 135 120 512 True 33,753 ± 8,349 / 5,937 ± 1,946 4.33 82.91 ± 0.89 / 83.43 ± 0.87 1.67 ± 2.22 / 46.20 ± 3.31 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
sentence-transformers/quora-distilbert-multilingual 135 120 512 True 26,458 ± 5,992 / 5,274 ± 1,731 4.33 82.91 ± 0.89 / 83.43 ± 0.87 1.67 ± 2.22 / 46.20 ± 3.31 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
KBLab/bert-base-swedish-cased 125 50 512 True 16,164 ± 2,392 / 4,574 ± 1,478 4.35 79.99 ± 1.13 / 80.76 ± 1.10 1.32 ± 2.66 / 49.75 ± 1.50 0.10 ± 0.15 / 0.13 ± 0.19 12.7.0 12.7.0 13.0.0
sentence-transformers/paraphrase-multilingual-mpnet-base-v2 278 250 512 True 15,100 ± 3,019 / 3,369 ± 1,103 4.35 81.70 ± 0.58 / 82.18 ± 0.53 0.25 ± 2.87 / 44.81 ± 2.90 0.00 ± 0.00 / 0.00 ± 0.00 12.7.0 0.0.0 13.0.0
dbmdz/bert-medium-historic-multilingual-cased 42 32 512 True 24,291 ± 4,887 / 5,096 ± 1,655 4.37 80.58 ± 0.45 / 81.29 ± 0.46 1.58 ± 2.34 / 49.16 ± 2.33 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
sentence-transformers/distiluse-base-multilingual-cased-v2 135 120 512 True 33,247 ± 8,123 / 6,017 ± 1,977 4.37 77.00 ± 1.08 / 77.85 ± 1.05 4.09 ± 2.27 / 47.26 ± 3.09 0.00 ± 0.00 / 0.00 ± 0.00 12.6.1 0.0.0 13.0.0
sentence-transformers/distiluse-base-multilingual-cased 135 120 512 True 19,206 ± 4,451 / 3,658 ± 1,187 4.37 77.16 ± 0.93 / 78.00 ± 0.91 4.09 ± 2.27 / 47.26 ± 3.09 0.00 ± 0.00 / 0.00 ± 0.00 12.6.1 0.0.0 13.0.0
sentence-transformers/paraphrase-xlm-r-multilingual-v1 278 250 512 True 20,154 ± 4,438 / 3,890 ± 1,256 4.37 80.92 ± 0.59 / 81.63 ± 0.56 1.19 ± 2.37 / 43.13 ± 4.70 0.00 ± 0.00 / 0.00 ± 0.00 12.6.1 0.0.0 13.0.0
DDSC/roberta-base-scandinavian 124 50 512 True 14,491 ± 2,800 / 3,182 ± 1,026 4.39 63.86 ± 20.89 / 64.42 ± 21.07 0.73 ± 2.11 / 49.36 ± 1.78 10.57 ± 1.30 / 15.52 ± 1.88 0.0.0 0.0.0 13.0.0
ibm-granite/granite-3b-code-base-2k (few-shot) 3483 49 2048 True 2,732 ± 868 / 662 ± 238 4.39 60.88 ± 2.03 / 60.02 ± 1.84 -0.35 ± 1.18 / 34.50 ± 1.25 18.54 ± 1.93 / 29.79 ± 1.35 13.0.0 13.0.0 13.0.0
FacebookAI/roberta-base 124 50 512 True 13,354 ± 3,334 / 2,451 ± 777 4.40 81.78 ± 0.95 / 82.55 ± 0.90 -1.18 ± 2.36 / 37.68 ± 3.64 0.69 ± 0.78 / 1.03 ± 1.15 0.0.0 0.0.0 13.0.0
sentence-transformers/distiluse-base-multilingual-cased-v1 135 120 512 True 34,042 ± 8,482 / 5,951 ± 1,950 4.40 75.63 ± 1.41 / 76.23 ± 1.36 3.65 ± 1.93 / 48.34 ± 1.78 0.00 ± 0.00 / 0.00 ± 0.00 12.6.1 0.0.0 13.0.0
pdelobelle/robbert-v2-dutch-base 116 40 512 True 15,481 ± 2,820 / 3,708 ± 1,186 4.42 78.59 ± 1.16 / 79.35 ± 1.08 0.65 ± 2.63 / 47.40 ± 3.23 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
senseable/WestLake-7B-v2 (few-shot) 7242 32 32768 False 5,993 ± 1,028 / 1,742 ± 561 4.43 67.42 ± 2.21 / 60.87 ± 3.22 7.02 ± 1.56 / 49.32 ± 2.91 0.65 ± 0.17 / 7.03 ± 0.26 12.6.1 12.6.1 13.0.0
dbmdz/bert-mini-historic-multilingual-cased 12 32 512 True 47,122 ± 9,661 / 9,714 ± 3,152 4.47 71.10 ± 6.05 / 71.10 ± 6.05 2.29 ± 2.11 / 48.81 ± 1.33 0.00 ± 0.00 / 0.00 ± 0.00 12.10.1 0.0.0 13.0.0
dbmdz/bert-tiny-historic-multilingual-cased 5 32 512 True 78,027 ± 15,466 / 17,064 ± 5,335 4.47 72.08 ± 1.25 / 72.93 ± 1.20 2.65 ± 2.40 / 48.48 ± 1.87 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
3ebdola/Dialectal-Arabic-XLM-R-Base 277 250 512 True 12,783 ± 2,537 / 2,712 ± 885 4.51 73.34 ± 1.57 / 74.24 ± 1.56 0.97 ± 2.25 / 44.25 ± 3.12 0.00 ± 0.00 / 0.01 ± 0.02 12.6.1 12.6.1 13.0.0
KBLab/albert-base-swedish-cased-alpha 14 50 512 True 15,925 ± 2,281 / 4,780 ± 1,554 4.51 73.80 ± 0.98 / 74.58 ± 0.92 0.81 ± 2.90 / 48.11 ± 2.18 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
ltg/norbert2 125 50 512 True 15,523 ± 2,863 / 3,690 ± 1,195 4.54 60.57 ± 0.86 / 60.42 ± 0.90 4.16 ± 2.63 / 47.13 ± 3.43 7.84 ± 0.99 / 14.80 ± 1.25 0.0.0 0.0.0 13.0.0
Maltehb/aelaectra-danish-electra-small-uncased 14 32 128 True 5,995 ± 135 / 3,839 ± 1,247 4.55 62.07 ± 1.18 / 61.72 ± 1.18 5.11 ± 3.80 / 47.64 ± 4.52 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
EuropeanParliament/EUBERT 94 66 512 True 20,070 ± 3,977 / 4,400 ± 1,435 4.59 59.50 ± 1.56 / 58.98 ± 1.55 3.25 ± 2.04 / 49.77 ± 1.40 5.56 ± 1.52 / 10.14 ± 2.41 12.6.1 12.6.1 13.0.0
ibm-granite/granite-3b-code-instruct-2k (few-shot) 3483 49 2048 True 9,059 ± 1,947 / 2,201 ± 728 4.61 56.88 ± 2.51 / 57.68 ± 2.14 -0.21 ± 2.20 / 34.20 ± 1.19 13.72 ± 2.56 / 24.11 ± 2.40 13.0.0 13.0.0 13.0.0
sarnikowski/convbert-medium-small-da-cased 24 29 512 True 13,821 ± 2,209 / 3,547 ± 1,184 4.61 59.66 ± 0.63 / 59.39 ± 0.60 4.58 ± 3.90 / 50.83 ± 2.59 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
sarnikowski/convbert-small-da-cased 13 29 512 True 14,273 ± 2,312 / 3,555 ± 1,187 4.65 58.50 ± 0.63 / 58.33 ± 0.70 5.96 ± 2.04 / 51.99 ± 1.53 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
ibm-granite/granite-3.0-3b-a800m-instruct (few-shot) 3374 49 4096 True 10,246 ± 3,021 / 1,629 ± 550 4.66 45.56 ± 1.92 / 42.78 ± 1.78 0.92 ± 2.19 / 40.29 ± 3.87 20.82 ± 1.22 / 31.49 ± 1.49 13.0.0 13.0.0 13.0.0
Maltehb/aelaectra-danish-electra-small-cased 14 32 512 True 4,593 ± 114 / 3,034 ± 973 4.76 58.52 ± 3.91 / 59.67 ± 3.83 1.09 ± 2.20 / 33.70 ± 0.56 0.00 ± 0.00 / 0.00 ± 0.00 12.7.0 12.7.0 13.0.0
ibm-granite/granite-3.0-3b-a800m-base (few-shot) 3374 49 4096 True 10,504 ± 3,028 / 1,678 ± 559 4.81 41.27 ± 4.31 / 40.85 ± 4.31 -0.20 ± 2.17 / 43.01 ± 3.39 19.69 ± 2.66 / 29.77 ± 2.80 13.0.0 13.0.0 13.0.0
fresh-xlm-roberta-base 277 250 512 True 2,214 ± 94 / 1,494 ± 229 4.92 48.70 ± 1.57 / 48.51 ± 1.57 2.37 ± 2.06 / 37.68 ± 3.25 0.00 ± 0.00 / 0.00 ± 0.00 0.0.0 0.0.0 13.0.0
HuggingFaceTB/SmolLM2-1.7B-Instruct (few-shot) 1711 49 8192 True 15,971 ± 3,654 / 3,609 ± 1,197 4.99 40.28 ± 3.29 / 42.15 ± 2.85 -0.26 ± 1.92 / 38.29 ± 4.03 10.68 ± 1.75 / 17.52 ± 1.85 13.1.0 13.1.0 13.1.0
HuggingFaceTB/SmolLM2-1.7B (few-shot) 1711 49 8192 True 16,249 ± 3,690 / 3,689 ± 1,226 5.15 27.91 ± 4.97 / 30.98 ± 4.28 -0.48 ± 0.89 / 33.97 ± 1.00 16.56 ± 1.66 / 26.11 ± 1.42 13.1.0 13.1.0 13.1.0
HuggingFaceTB/SmolLM2-135M (few-shot) 135 49 8192 True 26,346 ± 7,812 / 4,082 ± 1,372 5.37 25.51 ± 2.40 / 26.43 ± 1.77 0.46 ± 1.43 / 36.19 ± 3.30 0.15 ± 0.20 / 3.01 ± 0.44 13.1.0 13.1.0 13.1.0
HuggingFaceTB/SmolLM2-360M-Instruct (few-shot) 362 49 8192 True 21,777 ± 6,115 / 3,617 ± 1,211 5.41 26.85 ± 3.99 / 27.35 ± 3.60 -0.12 ± 1.62 / 45.84 ± 3.59 1.39 ± 0.54 / 4.02 ± 0.64 13.1.0 13.1.0 13.1.0
HuggingFaceTB/SmolLM2-360M (few-shot) 362 49 8192 True 22,023 ± 6,203 / 3,675 ± 1,231 5.41 28.14 ± 2.42 / 28.12 ± 2.55 -0.06 ± 0.67 / 33.85 ± 0.68 2.43 ± 0.73 / 6.05 ± 0.59 13.1.0 13.1.0 13.1.0
HuggingFaceTB/SmolLM2-135M-Instruct (few-shot) 135 49 8192 True 25,602 ± 7,583 / 3,953 ± 1,325 5.47 23.22 ± 3.22 / 24.30 ± 2.81 0.41 ± 2.39 / 39.64 ± 3.47 0.54 ± 0.33 / 3.34 ± 0.61 13.1.0 13.1.0 13.1.0
ssmits/Falcon2-5.5B-multilingual (few-shot) 5465 65 8192 True 7,692 ± 1,423 / 1,960 ± 644 6.01 0.00 ± 0.00 / 0.00 ± 0.00 0.00 ± 0.00 / 33.40 ± 0.34 0.00 ± 0.00 / 0.00 ± 0.00 13.0.0 13.0.0 13.0.0