Icelandic NLG 🇮🇸

gpt-4o-2024-05-13 (few-shot, val)

unknown

200

128000

True

916 ± 329 / 114 ± 38

1.15

81.19 ± 2.45 / 54.02 ± 5.60

49.86 ± 5.14 / 64.51 ± 3.68

51.10 ± 5.09 / 73.25 ± 3.42

29.64 ± 2.12 / 55.46 ± 1.12

68.25 ± 0.27 / 19.22 ± 0.51

91.27 ± 1.41 / 93.40 ± 1.09

70.85 ± 5.98 / 85.55 ± 3.05

12.10.0

13.2.0

12.10.2

12.10.0

12.10.8

12.10.2

gpt-4-1106-preview (few-shot, val)

unknown

100

128000

True

576 ± 221 / 81 ± 28

1.17

86.37 ± 1.19 / 82.25 ± 2.73

49.59 ± 2.76 / 62.50 ± 2.08

43.03 ± 5.07 / 71.18 ± 2.64

37.26 ± 2.60 / 66.04 ± 1.95

69.61 ± 0.61 / 23.98 ± 1.17

89.09 ± 1.59 / 91.76 ± 1.19

72.03 ± 3.91 / 86.09 ± 1.96

12.5.1

13.2.0

12.10.2

12.5.1

12.10.8

12.10.2

meta-llama/Llama-3.1-8B-Instruct (few-shot)

8030

128

131073

True

2,411 ± 465 / 686 ± 215

3.63

55.09 ± 1.76 / 40.34 ± 2.34

39.03 ± 1.68 / 56.67 ± 2.03

8.45 ± 1.33 / 51.00 ± 1.95

29.65 ± 1.09 / 56.96 ± 1.35

68.27 ± 0.25 / 22.14 ± 0.62

30.33 ± 1.72 / 47.81 ± 1.31

10.48 ± 2.28 / 54.89 ± 1.49

13.0.0

13.3.0

13.0.0

meta-llama/Llama-3.1-8B (few-shot)

8030

128

131073

True

2,439 ± 459 / 703 ± 219

3.71

52.57 ± 2.04 / 36.81 ± 2.12

44.00 ± 2.20 / 61.61 ± 1.71

4.87 ± 2.28 / 40.95 ± 4.97

30.12 ± 4.65 / 57.81 ± 4.78

66.17 ± 0.86 / 18.71 ± 0.89

25.93 ± 1.51 / 44.54 ± 1.14

8.48 ± 2.66 / 53.76 ± 1.54

12.11.0

13.3.0

12.11.0

13.0.0

mistralai/Ministral-8B-Instruct-2410 (few-shot)

8020

131

32768

True

5,821 ± 1,090 / 1,561 ± 506

3.73

56.09 ± 3.29 / 42.39 ± 2.51

39.51 ± 2.31 / 57.84 ± 2.34

8.51 ± 1.55 / 50.18 ± 2.85

29.23 ± 1.41 / 57.06 ± 0.62

65.90 ± 0.71 / 18.54 ± 0.92

30.58 ± 1.40 / 47.95 ± 1.05

6.77 ± 2.10 / 53.92 ± 1.32

13.0.0

13.3.0

13.0.0

meta-llama/Meta-Llama-3-8B (few-shot)

8030

128

8192

True

4,687 ± 1,121 / 967 ± 313

3.74

48.70 ± 3.02 / 34.52 ± 2.66

38.90 ± 3.77 / 57.62 ± 2.79

7.49 ± 2.51 / 43.40 ± 4.41

29.56 ± 5.47 / 55.53 ± 5.79

66.34 ± 1.09 / 19.13 ± 0.96

26.78 ± 1.59 / 45.17 ± 1.12

7.41 ± 3.26 / 52.13 ± 1.97

12.6.1

13.3.0

12.6.1

12.10.8

12.6.1

meta-llama/Meta-Llama-3-8B-Instruct (few-shot)

8030

128

8192

True

4,909 ± 1,215 / 978 ± 319

3.85

61.69 ± 2.17 / 41.25 ± 3.12

34.95 ± 4.34 / 51.53 ± 4.46

6.10 ± 1.61 / 48.74 ± 3.05

31.52 ± 2.08 / 58.96 ± 1.57

66.98 ± 1.04 / 19.84 ± 1.97

25.16 ± 1.55 / 43.57 ± 1.26

1.50 ± 1.22 / 48.33 ± 1.21

12.6.1

13.3.0

12.6.1

12.10.8

12.6.1

mistralai/Mistral-7B-v0.3 (few-shot)

7248

33

32768

True

4,120 ± 976 / 926 ± 306

4.28

44.68 ± 3.50 / 36.20 ± 4.20

31.17 ± 3.39 / 45.66 ± 3.73

0.12 ± 1.68 / 35.09 ± 1.17

25.52 ± 5.24 / 49.15 ± 5.21

61.40 ± 2.38 / 14.90 ± 1.60

10.25 ± 1.54 / 32.81 ± 1.22

5.24 ± 1.65 / 52.80 ± 2.41

12.10.4

13.3.0

12.10.4

12.10.5

12.10.8

12.10.4

mhenrichsen/hestenettetLM (few-shot)

7242

32

32768

True

5,160 ± 804 / 1,654 ± 516

4.29

50.82 ± 2.72 / 40.35 ± 4.51

26.12 ± 4.52 / 46.82 ± 4.45

0.99 ± 1.54 / 39.38 ± 3.81

25.74 ± 5.44 / 49.45 ± 5.29

61.72 ± 3.16 / 16.00 ± 1.82

10.78 ± 1.15 / 33.40 ± 0.91

3.94 ± 2.97 / 54.60 ± 1.62

12.5.2

13.3.0

12.3.2

12.10.8

12.3.2

mistralai/Mistral-7B-v0.1 (few-shot)

7242

32

32768

True

2,657 ± 524 / 880 ± 278

4.31

47.24 ± 2.54 / 37.77 ± 3.87

26.07 ± 4.82 / 46.35 ± 4.70

1.35 ± 1.70 / 39.37 ± 3.87

25.70 ± 5.36 / 49.31 ± 5.21

61.96 ± 3.10 / 16.11 ± 1.80

10.25 ± 0.96 / 32.89 ± 0.83

1.99 ± 2.95 / 54.48 ± 1.27

9.1.2

13.3.0

9.1.2

12.5.1

11.0.0

12.10.8

12.1.0

Nexusflow/Starling-LM-7B-beta (few-shot)

7242

32

8192

False

5,876 ± 1,021 / 1,677 ± 546

4.33

49.20 ± 2.64 / 40.79 ± 4.46

28.82 ± 2.16 / 48.33 ± 2.73

4.45 ± 1.40 / 51.11 ± 0.87

24.61 ± 3.36 / 54.99 ± 2.36

63.74 ± 2.25 / 18.29 ± 1.40

8.45 ± 1.35 / 31.54 ± 1.03

1.14 ± 0.97 / 50.10 ± 0.82

12.5.2

13.3.0

12.5.2

12.10.8

12.5.2

CohereForAI/aya-expanse-8b (few-shot)

8028

256

8192

False

5,581 ± 1,066 / 1,471 ± 483

4.49

28.98 ± 2.63 / 21.75 ± 1.89

19.83 ± 4.76 / 41.64 ± 3.64

4.93 ± 1.06 / 49.69 ± 2.65

24.72 ± 2.22 / 54.41 ± 1.43

63.45 ± 1.92 / 16.81 ± 2.23

10.97 ± 1.13 / 33.51 ± 0.81

4.23 ± 1.80 / 49.31 ± 0.84

13.0.0

13.3.0

13.0.0

senseable/WestLake-7B-v2 (few-shot)

7242

32

32768

False

5,993 ± 1,028 / 1,742 ± 561

4.52

56.71 ± 1.98 / 46.71 ± 5.28

7.92 ± 3.63 / 32.79 ± 2.05

3.44 ± 2.02 / 50.18 ± 1.14

21.55 ± 2.79 / 54.79 ± 2.02

65.39 ± 0.80 / 18.24 ± 1.00

9.11 ± 0.92 / 32.06 ± 0.70

3.30 ± 2.81 / 44.40 ± 1.61

12.6.1

13.3.0

12.6.1

12.10.8

12.6.1

mistralai/Mistral-7B-Instruct-v0.2 (few-shot)

7242

32

32768

False

634 ± 179 / 110 ± 35

4.54

43.11 ± 2.23 / 29.34 ± 3.27

22.55 ± 4.07 / 46.66 ± 3.12

3.40 ± 1.87 / 48.75 ± 1.47

19.18 ± 3.69 / 49.62 ± 2.59

65.01 ± 1.51 / 18.34 ± 1.35

5.49 ± 1.98 / 28.73 ± 1.39

0.24 ± 0.71 / 38.95 ± 0.84

9.2.0

13.3.0

9.3.1

12.4.0

12.10.8

12.1.0

timpal0l/Mistral-7B-v0.1-flashback-v2 (few-shot)

7242

32

32768

True

5,054 ± 1,200 / 1,056 ± 339

4.55

36.47 ± 4.24 / 30.33 ± 3.70

16.08 ± 4.33 / 34.46 ± 4.77

2.54 ± 1.29 / 50.66 ± 0.62

18.66 ± 4.26 / 38.73 ± 3.66

63.68 ± 1.75 / 16.38 ± 1.24

5.12 ± 1.30 / 28.85 ± 0.99

8.30 ± 1.28 / 57.35 ± 0.75

12.5.3

13.3.0

12.5.3

12.10.8

12.5.3

meta-llama/Llama-3.2-3B (few-shot)

3213

128

131073

True

3,713 ± 877 / 836 ± 267

4.59

42.04 ± 3.53 / 25.31 ± 1.59

14.86 ± 5.20 / 32.90 ± 5.58

0.00 ± 0.00 / 32.97 ± 0.29

24.07 ± 4.63 / 49.31 ± 4.59

61.80 ± 2.80 / 14.50 ± 2.00

13.79 ± 1.55 / 35.23 ± 1.26

0.25 ± 2.45 / 48.10 ± 1.67

13.0.0

13.3.0

13.0.0

occiglot/occiglot-7b-eu5-instruct (few-shot)

7242

32

32768

False

2,088 ± 352 / 706 ± 214

4.63

40.71 ± 2.93 / 34.57 ± 4.02

14.70 ± 7.78 / 36.09 ± 6.06

0.71 ± 2.00 / 36.90 ± 2.10

20.66 ± 3.67 / 45.91 ± 3.45

65.25 ± 0.97 / 19.09 ± 1.05

5.35 ± 1.32 / 28.11 ± 1.13

0.35 ± 2.49 / 51.16 ± 2.74

12.5.2

13.3.0

12.3.1

12.4.0

12.10.8

12.3.1

CohereForAI/aya-23-8B (few-shot)

8028

256

8192

False

5,608 ± 1,062 / 1,472 ± 479

4.64

47.16 ± 2.83 / 38.60 ± 4.04

19.77 ± 6.04 / 34.40 ± 4.93

3.84 ± 1.27 / 40.06 ± 3.79

21.75 ± 2.21 / 48.25 ± 2.02

59.16 ± 2.91 / 14.91 ± 1.85

3.70 ± 1.01 / 27.71 ± 0.66

-3.24 ± 3.16 / 49.96 ± 2.38

13.0.0

13.3.0

13.0.0

01-ai/Yi-1.5-6B (few-shot)

6061

64

4097

True

2,867 ± 550 / 793 ± 253

4.66

38.15 ± 2.30 / 28.10 ± 4.65

10.58 ± 5.45 / 31.23 ± 5.41

0.98 ± 1.44 / 35.86 ± 1.12

20.39 ± 3.04 / 44.20 ± 2.72

61.23 ± 2.31 / 14.45 ± 1.79

7.59 ± 1.44 / 31.07 ± 1.05

3.44 ± 2.28 / 46.89 ± 1.02

13.0.0

13.3.0

13.0.0

AI-Sweden-Models/gpt-sw3-6.7b-v2-instruct (few-shot)

7111

64

2048

True

2,383 ± 451 / 718 ± 221

4.67

19.39 ± 1.31 / 19.04 ± 1.36

33.82 ± 2.29 / 50.46 ± 3.81

0.01 ± 1.49 / 34.61 ± 0.73

20.92 ± 3.41 / 51.75 ± 2.10

66.55 ± 0.34 / 18.80 ± 0.51

6.02 ± 1.50 / 29.00 ± 1.28

-2.37 ± 1.71 / 39.83 ± 3.57

12.7.0

13.3.0

12.7.0

12.4.0

12.10.8

12.7.0

MaLA-LM/emma-500-llama2-7b (few-shot)

6738

32

4096

True

6,275 ± 1,193 / 1,755 ± 578

4.68

31.81 ± 1.93 / 29.47 ± 2.02

19.61 ± 6.74 / 33.96 ± 6.93

3.63 ± 1.69 / 44.49 ± 3.89

16.72 ± 7.29 / 46.83 ± 5.93

58.72 ± 3.28 / 13.71 ± 1.35

12.62 ± 1.36 / 34.51 ± 1.27

3.43 ± 2.18 / 44.56 ± 1.09

13.0.0

13.3.0

13.0.0

ibm-granite/granite-3.0-8b-instruct (few-shot)

8171

49

4096

True

5,090 ± 937 / 1,423 ± 466

4.72

42.67 ± 3.13 / 35.40 ± 3.93

9.95 ± 3.78 / 30.20 ± 5.91

1.11 ± 1.45 / 34.61 ± 0.82

22.25 ± 2.78 / 49.58 ± 2.19

63.81 ± 1.05 / 16.07 ± 1.19

5.12 ± 1.20 / 27.91 ± 0.90

0.89 ± 2.37 / 53.27 ± 1.64

13.0.0

13.3.0

13.0.0

utter-project/EuroLLM-9B-Instruct (few-shot)

9152

128

4096

False

5,763 ± 1,178 / 1,390 ± 446

4.72

24.92 ± 3.54 / 23.09 ± 3.30

9.76 ± 7.16 / 26.68 ± 6.80

0.15 ± 0.29 / 33.01 ± 0.31

28.18 ± 1.51 / 54.08 ± 0.71

64.67 ± 1.48 / 18.74 ± 1.31

11.10 ± 0.78 / 33.53 ± 0.69

0.89 ± 1.86 / 51.27 ± 1.18

13.1.0

13.3.0

13.1.0

HPLT/gpt-7b-nordic-prerelease (few-shot)

7550

131

4096

True

5,404 ± 931 / 1,638 ± 542

4.74

27.96 ± 3.08 / 25.78 ± 3.20

29.79 ± 6.62 / 43.62 ± 8.47

-0.00 ± 1.28 / 35.53 ± 1.87

23.17 ± 2.78 / 44.72 ± 2.82

55.57 ± 4.13 / 9.41 ± 1.58

0.94 ± 1.26 / 22.66 ± 0.57

-2.72 ± 3.17 / 53.79 ± 1.42

12.5.2

13.3.0

12.3.2

12.10.8

12.3.2

meta-llama/Llama-3.2-3B-Instruct (few-shot)

3213

128

131073

False

10,424 ± 2,641 / 2,081 ± 666

4.75

27.57 ± 1.71 / 22.52 ± 1.18

10.07 ± 5.45 / 28.32 ± 4.30

-1.39 ± 1.30 / 34.40 ± 1.96

22.98 ± 2.48 / 50.74 ± 1.59

62.00 ± 1.89 / 14.40 ± 1.93

13.33 ± 1.23 / 35.42 ± 0.93

0.74 ± 1.09 / 50.92 ± 0.73

13.0.0

13.3.0

13.0.0

neuralmagic/Sparse-Llama-3.1-8B-2of4 (few-shot)

8030

128

131072

True

2,841 ± 509 / 865 ± 263

4.75

17.32 ± 1.02 / 17.78 ± 0.81

27.63 ± 3.17 / 41.71 ± 3.27

-0.29 ± 1.38 / 33.12 ± 0.35

24.32 ± 3.61 / 47.17 ± 3.53

56.87 ± 4.23 / 13.32 ± 2.79

10.62 ± 1.58 / 33.06 ± 1.22

1.78 ± 1.49 / 55.89 ± 1.08

13.2.0

mistralai/Mistral-7B-Instruct-v0.1 (few-shot)

7242

32

32768

False

634 ± 179 / 110 ± 35

4.76

36.04 ± 2.59 / 24.74 ± 2.79

12.93 ± 5.42 / 30.40 ± 3.15

-0.36 ± 1.36 / 33.94 ± 0.32

18.06 ± 3.16 / 42.57 ± 2.89

62.80 ± 1.69 / 15.23 ± 1.01

5.44 ± 1.14 / 28.13 ± 1.06

6.35 ± 2.71 / 50.49 ± 1.57

9.3.1

13.3.0

9.3.1

12.4.0

12.10.8

12.1.0

occiglot/occiglot-7b-eu5 (few-shot)

7242

32

32768

True

2,219 ± 427 / 717 ± 224

4.78

40.08 ± 2.82 / 37.15 ± 4.07

15.49 ± 6.74 / 34.45 ± 5.64

1.59 ± 1.86 / 39.93 ± 4.19

15.98 ± 3.74 / 39.67 ± 3.36

62.55 ± 3.03 / 15.26 ± 2.31

5.98 ± 1.66 / 28.18 ± 1.30

-0.51 ± 1.95 / 47.23 ± 2.39

12.5.2

13.3.0

12.1.0

12.10.8

12.2.0

LumiOpen/Viking-7B (few-shot)

7550

131

4096

True

4,969 ± 1,109 / 1,134 ± 374

4.80

21.41 ± 5.01 / 19.94 ± 5.01

27.97 ± 6.85 / 41.29 ± 9.49

1.76 ± 1.62 / 42.86 ± 2.68

22.54 ± 1.74 / 44.93 ± 1.92

57.33 ± 2.62 / 10.47 ± 1.06

0.97 ± 1.29 / 26.73 ± 0.99

-0.36 ± 2.40 / 44.44 ± 1.08

12.7.0

13.3.0

12.7.0

12.10.8

12.7.0

ibm-granite/granite-8b-code-base-4k (few-shot)

8055

49

4096

True

2,313 ± 423 / 682 ± 210

4.80

50.89 ± 2.90 / 35.75 ± 4.67

9.18 ± 3.91 / 27.61 ± 4.83

0.50 ± 0.94 / 33.77 ± 0.30

17.43 ± 3.38 / 41.32 ± 3.18

59.94 ± 2.87 / 12.62 ± 1.91

5.52 ± 1.26 / 28.97 ± 0.97

1.73 ± 2.31 / 50.89 ± 3.35

13.0.0

13.3.0

13.0.0

meta-llama/Llama-2-7b-chat-hf (few-shot)

6738

32

4096

False

2,643 ± 455 / 800 ± 247

4.80

41.10 ± 3.35 / 40.54 ± 3.19

13.59 ± 6.27 / 36.87 ± 4.51

-1.07 ± 2.09 / 44.83 ± 2.20

16.13 ± 2.52 / 39.51 ± 1.98

62.30 ± 0.90 / 13.28 ± 1.36

3.16 ± 0.79 / 27.40 ± 0.79

1.84 ± 2.19 / 43.79 ± 0.73

9.3.1

13.3.0

9.3.1

12.4.0

12.10.8

12.1.0

bineric/NorskGPT-Llama-7B-v0.1 (few-shot)

6738

32

4096

False

5,384 ± 879 / 1,746 ± 553

4.82

34.62 ± 4.64 / 33.25 ± 4.37

22.29 ± 4.88 / 38.44 ± 3.78

-0.24 ± 1.43 / 33.75 ± 0.31

18.10 ± 1.85 / 43.52 ± 0.87

61.81 ± 0.98 / 15.04 ± 0.70

3.06 ± 1.01 / 28.01 ± 0.86

-1.90 ± 2.28 / 44.34 ± 1.19

12.5.2

13.3.0

12.3.2

12.10.8

12.3.2

ibm-granite/granite-3.0-8b-base (few-shot)

8171

49

4096

True

5,116 ± 943 / 1,436 ± 472

4.82

37.82 ± 3.63 / 32.97 ± 3.97

2.60 ± 2.91 / 15.68 ± 0.75

-0.12 ± 1.31 / 33.78 ± 0.32

21.59 ± 2.22 / 47.09 ± 1.09

62.35 ± 3.34 / 16.15 ± 1.66

6.54 ± 1.30 / 28.63 ± 1.24

1.83 ± 3.53 / 49.23 ± 2.07

13.0.0

13.3.0

13.0.0

microsoft/Phi-3-mini-4k-instruct (few-shot)

3821

32

4096

True

5,224 ± 1,371 / 1,063 ± 358

4.83

33.05 ± 4.29 / 29.75 ± 3.67

14.42 ± 3.18 / 37.82 ± 2.65

0.71 ± 1.18 / 34.80 ± 0.88

17.23 ± 2.51 / 39.88 ± 1.59

60.08 ± 1.45 / 13.80 ± 0.81

2.67 ± 1.35 / 27.04 ± 1.07

2.74 ± 2.28 / 53.18 ± 0.93

12.10.5

13.3.0

12.10.5

12.10.8

12.10.5

google/gemma-2-2b (few-shot)

2614

256

8193

True

5,235 ± 1,226 / 1,154 ± 366

4.84

10.67 ± 5.23 / 13.01 ± 3.26

21.00 ± 7.68 / 34.05 ± 7.03

-0.21 ± 0.99 / 33.71 ± 0.28

20.22 ± 3.01 / 43.48 ± 2.32

59.97 ± 2.49 / 13.32 ± 2.03

10.08 ± 1.82 / 32.08 ± 1.37

1.16 ± 2.48 / 51.00 ± 1.70

13.0.0

13.3.0

13.0.0

microsoft/Phi-3-mini-128k-instruct (few-shot)

3821

32

131072

True

7,312 ± 1,668 / 1,609 ± 525

4.84

27.22 ± 3.65 / 24.21 ± 2.67

17.80 ± 4.22 / 36.50 ± 4.92

1.31 ± 1.28 / 39.67 ± 4.39

17.24 ± 2.72 / 41.15 ± 1.57

62.00 ± 1.66 / 15.80 ± 1.12

1.85 ± 1.18 / 26.88 ± 0.86

1.06 ± 2.84 / 47.19 ± 2.14

12.9.1

13.3.0

12.9.1

12.10.0

12.10.8

12.10.0

google/gemma-2-2b-it (few-shot)

2614

256

8193

True

5,374 ± 1,233 / 1,193 ± 377

4.86

14.79 ± 4.70 / 13.03 ± 2.92

15.24 ± 6.48 / 34.96 ± 6.66

1.10 ± 1.54 / 34.67 ± 0.87

25.42 ± 1.09 / 51.72 ± 1.28

62.67 ± 1.62 / 16.91 ± 1.38

10.76 ± 1.56 / 32.34 ± 1.02

-5.20 ± 1.82 / 53.08 ± 0.91

13.0.0

13.3.0

13.0.0

meta-llama/Llama-2-7b-hf (few-shot)

6738

32

4096

True

930 ± 310 / 128 ± 43

4.87

32.71 ± 2.77 / 32.17 ± 2.13

13.24 ± 8.29 / 31.92 ± 6.23

0.66 ± 1.75 / 40.36 ± 4.19

18.04 ± 4.05 / 41.40 ± 3.27

60.73 ± 3.02 / 14.02 ± 1.57

3.65 ± 1.33 / 26.91 ± 1.00

-0.00 ± 2.41 / 44.93 ± 0.92

9.2.0

13.3.0

9.2.0

12.5.1

11.0.0

12.10.8

12.1.0

norallm/normistral-7b-warm-instruct (few-shot)

7248

33

4096

True

6,194 ± 949 / 1,967 ± 619

4.89

36.59 ± 3.56 / 27.50 ± 2.53

11.75 ± 4.91 / 34.26 ± 5.32

0.86 ± 2.41 / 36.44 ± 1.27

14.58 ± 2.13 / 37.44 ± 1.86

61.99 ± 1.16 / 15.07 ± 0.81

1.48 ± 1.63 / 23.23 ± 0.93

-0.98 ± 2.63 / 56.13 ± 0.62

12.7.0

13.3.0

12.7.0

12.10.8

12.7.0

tollefj/nordavind-7b-instruct-warm (few-shot)

7248

33

2048

False

6,450 ± 961 / 2,082 ± 658

4.92

34.76 ± 4.42 / 23.42 ± 2.33

12.07 ± 3.53 / 32.35 ± 5.40

0.77 ± 1.05 / 39.63 ± 2.41

12.80 ± 2.37 / 30.77 ± 2.12

61.23 ± 1.78 / 15.53 ± 0.95

2.01 ± 0.96 / 25.83 ± 1.02

-0.76 ± 3.69 / 53.64 ± 2.57

12.5.2

13.3.0

12.3.2

12.4.0

12.10.8

12.3.2

AI-Sweden-Models/gpt-sw3-356m-instruct (few-shot)

471

64

2048

True

5,855 ± 1,373 / 1,223 ± 391

4.96

17.79 ± 1.18 / 18.12 ± 1.18

21.09 ± 4.78 / 36.73 ± 5.45

0.08 ± 0.15 / 33.73 ± 0.26

15.04 ± 2.53 / 34.77 ± 1.72

59.45 ± 1.99 / 12.89 ± 1.04

1.06 ± 1.26 / 22.59 ± 0.63

5.69 ± 2.26 / 56.71 ± 0.87

12.7.0

13.3.0

12.7.0

12.10.8

12.7.0

meta-llama/Llama-3.2-1B-Instruct (few-shot)

1236

128

131073

False

7,436 ± 1,846 / 1,508 ± 479

4.98

33.76 ± 3.44 / 32.21 ± 2.82

6.91 ± 4.60 / 27.56 ± 4.17

-0.13 ± 1.34 / 42.66 ± 3.62

12.43 ± 2.62 / 33.38 ± 1.39

59.46 ± 2.42 / 13.23 ± 1.87

2.09 ± 1.34 / 25.68 ± 0.67

3.56 ± 2.07 / 56.72 ± 0.99

13.0.0

13.3.0

13.0.0

ibm-granite/granite-3.0-2b-instruct (few-shot)

2634

49

4097

True

10,194 ± 2,403 / 2,193 ± 731

5.00

30.21 ± 4.29 / 29.79 ± 3.81

5.88 ± 2.95 / 21.46 ± 3.02

3.67 ± 1.87 / 48.02 ± 3.06

15.12 ± 1.65 / 40.55 ± 1.13

56.75 ± 2.25 / 10.37 ± 1.58

3.22 ± 0.92 / 27.64 ± 0.84

1.88 ± 2.37 / 49.26 ± 1.60

13.0.0

13.3.0

13.0.0

HuggingFaceTB/SmolLM2-1.7B-Instruct (few-shot)

1711

49

8192

True

15,971 ± 3,654 / 3,609 ± 1,197

5.04

26.23 ± 3.53 / 23.26 ± 2.30

6.86 ± 5.25 / 23.96 ± 4.13

2.69 ± 1.42 / 45.61 ± 2.69

10.84 ± 2.59 / 28.57 ± 2.09

60.43 ± 2.70 / 15.28 ± 1.36

1.93 ± 1.38 / 25.06 ± 1.16

3.27 ± 2.97 / 54.02 ± 1.79

13.1.0

13.3.0

13.1.0

ibm-granite/granite-3b-code-base-2k (few-shot)

3483

49

2048

True

2,732 ± 868 / 662 ± 238

5.05

38.52 ± 3.25 / 28.84 ± 5.49

5.15 ± 4.46 / 21.10 ± 4.20

0.00 ± 0.00 / 33.69 ± 0.28

12.94 ± 2.60 / 36.06 ± 2.17

58.58 ± 4.00 / 12.91 ± 2.09

2.11 ± 1.02 / 27.80 ± 1.02

-4.75 ± 2.84 / 50.40 ± 4.21

13.0.0

13.3.0

13.0.0

Qwen/Qwen1.5-4B-Chat (few-shot)

3950

152

32768

False

4,347 ± 893 / 1,135 ± 365

5.06

25.65 ± 2.99 / 22.30 ± 2.30

7.56 ± 4.87 / 26.07 ± 4.12

-0.35 ± 2.01 / 44.36 ± 4.13

14.46 ± 2.66 / 32.31 ± 1.66

62.11 ± 2.22 / 14.98 ± 1.53

4.50 ± 1.47 / 28.86 ± 1.09

-1.89 ± 2.66 / 43.72 ± 0.92

12.5.2

13.3.0

12.1.0

12.5.2

12.1.0

12.10.8

12.1.0

ibm-granite/granite-8b-code-instruct-4k (few-shot)

8055

49

4096

True

5,617 ± 995 / 1,623 ± 540

5.10

43.20 ± 3.61 / 32.10 ± 3.76

2.54 ± 2.60 / 17.05 ± 2.70

0.00 ± 0.00 / 33.69 ± 0.28

14.28 ± 2.94 / 38.21 ± 2.38

49.66 ± 3.28 / 7.97 ± 1.43

3.07 ± 1.66 / 25.74 ± 1.09

2.79 ± 3.13 / 47.97 ± 1.43

13.0.0

13.3.0

13.0.0

AI-Sweden-Models/gpt-sw3-1.3b (few-shot)

1445

64

2048

True

4,608 ± 988 / 1,115 ± 354

5.11

1.42 ± 1.60 / 3.11 ± 1.85

3.65 ± 4.07 / 18.36 ± 3.96

0.75 ± 0.73 / 45.87 ± 2.20

23.33 ± 2.22 / 45.28 ± 1.58

64.23 ± 1.78 / 15.08 ± 2.03

0.40 ± 1.88 / 23.35 ± 1.22

0.68 ± 4.15 / 50.85 ± 2.65

12.7.0

13.3.0

12.7.0

12.10.8

12.7.0

google/gemma-2b (few-shot)

2506

256

8192

True

6,087 ± 1,046 / 1,902 ± 563

5.12

8.83 ± 5.85 / 9.93 ± 4.70

16.00 ± 4.82 / 35.72 ± 5.59

0.31 ± 1.95 / 45.42 ± 3.51

16.08 ± 2.91 / 37.41 ± 2.44

60.00 ± 2.62 / 13.07 ± 1.31

2.52 ± 1.20 / 26.23 ± 1.40

0.00 ± 2.53 / 56.42 ± 0.98

12.5.2

13.3.0

12.1.0

12.10.8

12.1.0

HuggingFaceTB/SmolLM2-1.7B (few-shot)

1711

49

8192

True

16,249 ± 3,690 / 3,689 ± 1,226

5.16

20.50 ± 4.51 / 18.93 ± 3.84

8.08 ± 2.30 / 27.38 ± 2.88

0.83 ± 1.65 / 45.84 ± 2.02

10.84 ± 2.45 / 28.36 ± 1.48

57.52 ± 4.61 / 13.43 ± 2.11

3.16 ± 0.85 / 26.13 ± 0.67

-1.83 ± 3.11 / 49.80 ± 1.96

13.1.0

13.3.0

13.1.0

ibm-granite/granite-3.0-2b-base (few-shot)

2534

49

4097

True

10,187 ± 2,363 / 2,204 ± 737

5.16

29.51 ± 2.44 / 29.14 ± 2.34

3.55 ± 1.67 / 15.76 ± 0.47

-0.32 ± 0.63 / 33.70 ± 0.28

12.36 ± 2.12 / 34.06 ± 1.47

59.43 ± 1.83 / 13.09 ± 1.58

2.35 ± 1.26 / 27.85 ± 0.87

0.01 ± 2.73 / 45.18 ± 1.36

13.0.0

13.3.0

13.0.0

Qwen/Qwen1.5-4B (few-shot)

3950

152

32768

True

3,248 ± 739 / 761 ± 252

5.19

15.66 ± 5.89 / 15.78 ± 3.95

5.20 ± 3.94 / 22.73 ± 2.93

-0.55 ± 1.06 / 39.57 ± 3.61

14.11 ± 3.08 / 34.56 ± 2.38

57.17 ± 3.07 / 11.73 ± 1.00

5.46 ± 1.45 / 29.23 ± 1.11

-1.71 ± 3.79 / 50.88 ± 1.29

12.5.2

13.3.0

12.1.0

12.10.8

12.1.0

ibm-granite/granite-3.0-3b-a800m-instruct (few-shot)

3374

49

4096

True

10,246 ± 3,021 / 1,629 ± 550

5.19

23.14 ± 2.08 / 23.09 ± 2.24

5.07 ± 2.69 / 27.01 ± 2.31

0.18 ± 1.67 / 33.93 ± 0.34

14.15 ± 2.49 / 36.10 ± 1.65

60.80 ± 0.81 / 10.79 ± 1.85

2.86 ± 1.31 / 26.61 ± 1.04

-1.31 ± 2.24 / 50.29 ± 1.93

13.0.0

13.3.0

13.0.0

01-ai/Yi-6B (few-shot)

6061

64

4096

True

6,435 ± 1,316 / 1,632 ± 549

5.20

0.00 ± 0.00 / 0.00 ± 0.00

8.76 ± 4.64 / 27.75 ± 4.75

2.12 ± 1.40 / 38.45 ± 2.47

16.91 ± 2.57 / 40.63 ± 2.83

60.02 ± 3.15 / 14.22 ± 1.52

4.35 ± 1.25 / 28.94 ± 1.09

0.72 ± 2.33 / 52.54 ± 2.18

9.3.2

13.3.0

10.0.0

12.5.1

12.0.0

12.10.8

12.1.0

google/gemma-2b-it (few-shot)

2506

256

8192

False

6,471 ± 1,142 / 1,961 ± 584

5.20

20.49 ± 2.30 / 18.33 ± 1.40

7.64 ± 3.36 / 28.09 ± 4.00

-0.01 ± 2.13 / 46.02 ± 2.71

10.95 ± 2.39 / 37.64 ± 0.75

59.16 ± 0.96 / 9.92 ± 1.05

0.45 ± 1.44 / 22.94 ± 0.76

0.62 ± 1.42 / 56.02 ± 0.95

12.5.2

13.3.0

12.1.0

12.4.0

12.10.8

12.1.0

timpal0l/Mistral-7B-v0.1-flashback-v2-instruct (few-shot)

7242

32

32768

False

5,172 ± 813 / 1,647 ± 518

5.24

24.98 ± 5.71 / 25.35 ± 4.78

13.57 ± 4.68 / 28.38 ± 5.14

1.18 ± 1.09 / 39.01 ± 2.76

8.52 ± 2.30 / 21.32 ± 2.25

39.94 ± 9.39 / 5.18 ± 1.53

4.83 ± 1.40 / 29.14 ± 1.10

4.70 ± 2.96 / 56.56 ± 0.97

12.5.2

13.3.0

12.3.2

12.10.8

12.3.2

Qwen/Qwen1.5-1.8B-Chat (few-shot)

1837

152

32768

False

8,304 ± 1,846 / 1,933 ± 617

5.29

14.15 ± 1.92 / 14.96 ± 2.11

6.87 ± 6.27 / 28.55 ± 3.29

0.78 ± 1.70 / 44.74 ± 3.57

7.80 ± 1.32 / 23.47 ± 1.64

57.27 ± 1.42 / 10.43 ± 0.97

1.62 ± 1.18 / 25.09 ± 1.06

1.92 ± 2.32 / 50.07 ± 2.68

12.5.2

13.3.0

12.1.0

12.5.0

12.10.8

12.1.0

meta-llama/Llama-3.2-1B (few-shot)

1236

128

131073

True

7,577 ± 1,884 / 1,555 ± 492

5.30

17.77 ± 6.81 / 16.42 ± 6.19

9.23 ± 3.97 / 26.61 ± 5.18

-0.35 ± 1.50 / 34.29 ± 1.64

8.15 ± 2.21 / 30.75 ± 2.05

54.21 ± 3.27 / 11.06 ± 1.56

0.71 ± 1.34 / 25.40 ± 0.83

2.88 ± 1.84 / 54.45 ± 2.60

13.0.0

13.3.0

13.0.0

utter-project/EuroLLM-1.7B-Instruct (few-shot)

1657

128

4096

True

15,009 ± 4,072 / 2,702 ± 878

5.31

13.29 ± 2.33 / 14.72 ± 2.11

6.83 ± 5.62 / 31.41 ± 4.70

0.92 ± 1.51 / 42.09 ± 3.39

7.49 ± 1.26 / 24.16 ± 1.95

56.76 ± 2.66 / 9.80 ± 1.05

0.59 ± 0.49 / 23.09 ± 0.56

1.87 ± 3.67 / 49.58 ± 3.20

13.1.0

13.3.0

13.1.0

Qwen/Qwen1.5-1.8B (few-shot)

1837

152

32768

True

5,666 ± 1,328 / 1,256 ± 408

5.32

12.26 ± 4.13 / 12.77 ± 3.60

5.43 ± 4.09 / 25.93 ± 4.06

0.94 ± 1.34 / 40.66 ± 3.73

6.31 ± 1.01 / 20.24 ± 2.02

55.32 ± 3.49 / 8.91 ± 1.05

3.65 ± 1.45 / 26.36 ± 0.92

1.13 ± 3.74 / 52.30 ± 2.26

12.5.2

13.3.0

12.1.0

12.10.8

12.1.0

ibm-granite/granite-3b-code-instruct-2k (few-shot)

3483

49

2048

True

9,059 ± 1,947 / 2,201 ± 728

5.34

33.57 ± 2.48 / 33.47 ± 2.48

0.60 ± 2.01 / 16.42 ± 1.16

0.00 ± 0.00 / 33.69 ± 0.28

11.27 ± 2.38 / 33.54 ± 1.64

49.32 ± 3.71 / 8.01 ± 2.04

1.37 ± 1.11 / 25.57 ± 1.36

-4.04 ± 3.58 / 50.55 ± 3.65

13.0.0

13.3.0

13.0.0

ibm-granite/granite-3.0-3b-a800m-base (few-shot)

3374

49

4096

True

10,504 ± 3,028 / 1,678 ± 559

5.36

18.07 ± 3.62 / 18.73 ± 2.54

1.67 ± 1.99 / 20.92 ± 3.22

-0.72 ± 1.22 / 33.96 ± 0.50

12.27 ± 2.81 / 30.66 ± 1.47

56.49 ± 3.33 / 8.77 ± 1.24

0.32 ± 1.35 / 24.82 ± 1.04

1.00 ± 2.57 / 51.21 ± 2.91

13.0.0

13.3.0

13.0.0

ibm-granite/granite-7b-instruct (few-shot)

6738

32

4096

True

3,136 ± 558 / 942 ± 290

5.38

19.07 ± 1.70 / 20.43 ± 1.80

-0.51 ± 3.34 / 18.63 ± 3.02

0.99 ± 1.27 / 39.24 ± 3.43

9.54 ± 0.94 / 25.84 ± 1.85

59.42 ± 0.72 / 10.23 ± 1.25

0.79 ± 1.16 / 24.04 ± 0.75

-3.20 ± 3.60 / 51.52 ± 1.92

13.2.0

HuggingFaceTB/SmolLM2-360M (few-shot)

362

49

8192

True

22,023 ± 6,203 / 3,675 ± 1,231

5.44

13.43 ± 1.36 / 13.81 ± 1.45

1.77 ± 3.85 / 25.49 ± 2.99

1.14 ± 1.52 / 36.93 ± 3.69

3.71 ± 1.14 / 16.21 ± 2.86

51.93 ± 3.96 / 8.48 ± 0.87

0.95 ± 1.48 / 22.52 ± 1.05

2.90 ± 2.91 / 56.72 ± 0.69

13.1.0

13.3.0

13.1.0

Qwen/Qwen1.5-0.5B (few-shot)

620

152

32768

True

11,371 ± 2,924 / 2,122 ± 692

5.44

16.20 ± 1.52 / 16.96 ± 1.71

1.95 ± 4.51 / 22.10 ± 3.03

-0.57 ± 1.20 / 41.25 ± 3.51

3.31 ± 0.82 / 16.86 ± 2.98

56.00 ± 3.13 / 10.05 ± 0.73

1.96 ± 1.72 / 25.55 ± 1.37

0.85 ± 1.91 / 52.12 ± 2.92

12.5.2

13.3.0

12.1.0

12.10.8

12.1.0

ibm-granite/granite-3.0-1b-a400m-instruct (few-shot)

1335

49

4096

True

7,964 ± 2,255 / 1,299 ± 433

5.45

17.74 ± 3.49 / 16.16 ± 1.89

3.84 ± 4.22 / 27.14 ± 5.28

-1.13 ± 1.09 / 44.10 ± 3.57

7.74 ± 1.43 / 24.78 ± 1.48

58.60 ± 1.15 / 10.95 ± 1.08

-1.11 ± 1.33 / 22.73 ± 0.90

-4.23 ± 3.18 / 51.91 ± 2.86

13.2.0

ibm-granite/granite-3.0-1b-a400m-base (few-shot)

1385

49

4096

True

7,808 ± 2,183 / 1,289 ± 428

5.50

5.62 ± 3.01 / 8.37 ± 2.88

4.82 ± 4.50 / 22.80 ± 5.16

-0.20 ± 0.76 / 33.73 ± 0.26

4.94 ± 1.33 / 18.58 ± 2.50

58.01 ± 2.21 / 10.21 ± 1.25

0.78 ± 1.22 / 22.17 ± 0.62

-2.14 ± 2.71 / 51.91 ± 3.81

13.2.0

Qwen/Qwen1.5-0.5B-Chat (few-shot)

620

152

32768

False

11,740 ± 3,000 / 2,209 ± 721

5.51

9.50 ± 3.17 / 9.41 ± 3.40

1.63 ± 1.45 / 19.19 ± 0.27

1.76 ± 1.62 / 38.51 ± 3.72

3.14 ± 0.71 / 17.84 ± 2.26

58.92 ± 1.57 / 10.09 ± 1.41

-1.28 ± 1.48 / 24.82 ± 0.92

1.48 ± 3.56 / 53.95 ± 2.45

12.5.2

13.3.0

12.1.0

12.5.0

12.10.8

12.1.0

HuggingFaceTB/SmolLM2-135M (few-shot)

135

49

8192

True

26,346 ± 7,812 / 4,082 ± 1,372

5.54

14.74 ± 2.42 / 16.01 ± 2.04

2.10 ± 3.52 / 23.56 ± 3.05

-0.25 ± 0.60 / 34.69 ± 3.02

1.35 ± 0.73 / 10.92 ± 2.52

52.66 ± 3.89 / 8.73 ± 0.71

1.21 ± 0.80 / 26.20 ± 1.03

1.69 ± 2.26 / 49.46 ± 3.56

13.1.0

13.3.0

13.1.0

HuggingFaceTB/SmolLM2-360M-Instruct (few-shot)

362

49

8192

True

21,777 ± 6,115 / 3,617 ± 1,211

5.54

13.60 ± 1.50 / 13.99 ± 1.39

3.12 ± 4.92 / 27.55 ± 4.21

0.28 ± 1.41 / 37.58 ± 4.34

4.09 ± 1.03 / 16.34 ± 2.86

50.00 ± 3.22 / 7.84 ± 0.71

-0.11 ± 0.95 / 22.28 ± 0.83

2.51 ± 3.01 / 51.41 ± 3.27

13.1.0

13.3.0

13.1.0

RJuro/kanelsnegl-v0.2 (few-shot)

7242

32

512

True

1,373 ± 120 / 709 ± 172

5.55

23.67 ± 5.16 / 23.19 ± 4.37

4.00 ± 2.53 / 25.28 ± 2.69

0.00 ± 0.00 / 33.69 ± 0.28

0.00 ± 0.00 / 14.61 ± 2.02

50.54 ± 0.14 / 3.11 ± 0.06

0.00 ± 0.00 / 22.04 ± 0.48

0.00 ± 0.00 / 56.52 ± 0.89

12.7.0

13.3.0

12.7.0

12.10.8

12.7.0

RuterNorway/Llama-2-7b-chat-norwegian (few-shot)

6738

32

4096

False

10,890 ± 2,686 / 2,186 ± 750

5.62

9.48 ± 1.48 / 10.10 ± 1.44

3.32 ± 3.76 / 21.42 ± 3.52

0.07 ± 1.06 / 43.54 ± 3.63

1.04 ± 0.96 / 7.35 ± 3.52

55.16 ± 1.26 / 10.52 ± 1.13

-0.80 ± 2.00 / 23.89 ± 0.86

-0.16 ± 0.86 / 32.02 ± 2.77

9.3.1

13.3.0

9.3.1

12.5.2

12.4.0

12.10.8

12.1.0

state-spaces/mamba-2.8b-hf (few-shot)

2768

50

32769

True

2,722 ± 495 / 766 ± 250

5.63

18.38 ± 1.54 / 20.76 ± 1.34

-1.70 ± 2.31 / 21.85 ± 2.75

0.49 ± 1.54 / 42.54 ± 4.23

6.30 ± 2.23 / 23.24 ± 0.99

51.62 ± 2.09 / 8.63 ± 0.96

2.25 ± 1.26 / 26.07 ± 1.15

-6.17 ± 1.44 / 52.90 ± 2.42

13.0.0

13.3.0

13.0.0

HuggingFaceTB/SmolLM2-135M-Instruct (few-shot)

135

49

8192

True

25,602 ± 7,583 / 3,953 ± 1,325

5.64

13.70 ± 2.05 / 15.01 ± 2.07

3.01 ± 2.55 / 22.58 ± 2.22

-0.83 ± 0.71 / 32.99 ± 0.27

0.94 ± 0.54 / 10.22 ± 2.52

50.30 ± 4.32 / 7.82 ± 0.77

1.10 ± 1.16 / 25.43 ± 1.07

-0.07 ± 3.01 / 46.28 ± 3.42

13.1.0

13.3.0

13.1.0

NorGLM/NorGPT-369M (few-shot)

unknown

64

2048

True

19,896 ± 5,099 / 3,848 ± 1,251

5.87

1.68 ± 1.40 / 1.54 ± 1.28

0.51 ± 2.62 / 23.22 ± 2.96

-1.38 ± 1.13 / 34.41 ± 2.16

0.08 ± 0.09 / 10.05 ± 2.08

44.02 ± 1.31 / 6.35 ± 0.43

0.15 ± 1.45 / 23.95 ± 1.29

0.28 ± 1.39 / 32.09 ± 2.15

12.5.2

13.3.0

12.5.2

12.10.8

12.5.2

Sigurdur/icebreaker (few-shot)

110

32

1024

False

48,619 ± 7,681 / 13,831 ± 4,404

5.88

0.00 ± 0.00 / 0.00 ± 0.00

0.00 ± 0.00 / 18.79 ± 0.22

0.00 ± 0.00 / 33.69 ± 0.28

0.00 ± 0.00 / 3.90 ± 0.28

44.80 ± 0.65 / 3.34 ± 0.08

0.23 ± 0.75 / 22.11 ± 0.48

0.38 ± 0.75 / 56.53 ± 0.89

12.5.2

13.3.0

12.5.2

12.10.8

12.5.2

Sigurdur/icechat (few-shot)

110

32

1024

False

49,558 ± 7,930 / 13,921 ± 4,425

5.93

0.00 ± 0.00 / 0.00 ± 0.00

0.00 ± 0.00 / 18.79 ± 0.22

0.00 ± 0.00 / 33.69 ± 0.28

0.00 ± 0.00 / 0.64 ± 0.34

42.46 ± 0.47 / 3.58 ± 0.45

0.00 ± 0.00 / 22.04 ± 0.48

0.00 ± 0.00 / 56.52 ± 0.89

12.5.2

13.3.0

12.5.2

12.10.8

12.5.2

ssmits/Falcon2-5.5B-multilingual (few-shot)

5465

65

8192

True

7,692 ± 1,423 / 1,960 ± 644

6.02

0.00 ± 0.00 / 0.00 ± 0.00

0.00 ± 0.00 / 18.79 ± 0.22

0.00 ± 0.00 / 33.69 ± 0.28

0.00 ± 0.00 / 0.02 ± 0.01

36.58 ± 2.96 / 0.16 ± 0.02

-0.00 ± 1.73 / 24.71 ± 1.42

0.00 ± 0.00 / 43.48 ± 0.89

13.0.0

13.3.0

13.0.0

ai-forever/mGPT (few-shot)

unknown

100

2048

True

11,734 ± 3,124 / 2,174 ± 720

6.23

0.00 ± 0.00 / 0.00 ± 0.00

3.82 ± 2.16 / 23.71 ± 2.49

0.00 ± 0.00 / 33.69 ± 0.28

0.00 ± 0.00 / 0.05 ± 0.03

17.11 ± 1.37 / 0.96 ± 0.09

-0.02 ± 1.16 / 22.75 ± 0.50

0.47 ± 4.14 / 46.93 ± 3.13

9.3.1

13.3.0

11.0.0

12.5.1

12.0.0

12.10.8

12.1.0

Sigurdur/jonas-hallgrimsson-gpt2 (few-shot)

125

51

512

False

32,644 ± 3,887 / 11,289 ± 3,585

6.53

0.00 ± 0.00 / 0.00 ± 0.00

0.00 ± 0.00 / 18.79 ± 0.22

0.00 ± 0.00 / 33.69 ± 0.28

0.00 ± 0.00 / 0.00 ± 0.00

-0.26 ± 0.90 / 22.02 ± 0.48

-0.01 ± 1.21 / 55.08 ± 0.99

12.5.2

13.3.0

12.5.2

12.10.8

12.5.2