English NLG 🇬🇧

gpt-4-0613 (few-shot, val)

unknown

100

8191

True

597 ± 197 / 93 ± 33

1.21

78.06 ± 2.73 / 70.76 ± 3.80

69.06 ± 2.20 / 76.04 ± 1.60

55.76 ± 3.84 / 76.34 ± 1.44

67.35 ± 1.98 / 85.76 ± 0.77

70.54 ± 0.24 / 26.98 ± 0.49

72.27 ± 2.36 / 78.75 ± 1.88

91.51 ± 2.15 / 93.59 ± 1.62

12.2.0

12.1.0

12.2.0

12.5.2

Qwen/QwQ-32B-Preview (few-shot)

32764

152

32768

True

2,258 ± 1,221 / 198 ± 67

1.22

76.84 ± 1.07 / 71.81 ± 0.77

68.94 ± 0.65 / 68.76 ± 1.24

57.74 ± 2.22 / 78.11 ± 1.52

71.22 ± 1.19 / 86.06 ± 0.60

68.96 ± 0.08 / 21.49 ± 0.18

75.15 ± 0.76 / 81.28 ± 0.58

87.36 ± 0.67 / 90.41 ± 0.52

14.0.4

meta-llama/Llama-3.1-405B-Instruct-FP8 (few-shot)

405869

128

131072

True

799 ± 246 / 112 ± 38

1.22

82.86 ± 0.87 / 81.12 ± 1.04

70.60 ± 1.14 / 73.47 ± 1.06

53.80 ± 1.81 / 76.34 ± 1.02

62.69 ± 1.45 / 82.32 ± 0.61

69.33 ± 0.08 / 23.72 ± 0.38

80.39 ± 0.78 / 85.32 ± 0.59

88.02 ± 0.99 / 91.00 ± 0.75

14.0.4

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

unknown

100

128254

True

576 ± 221 / 81 ± 28

1.23

81.79 ± 1.39 / 65.51 ± 4.16

67.55 ± 2.34 / 74.04 ± 1.80

51.21 ± 4.96 / 75.11 ± 2.42

66.60 ± 1.45 / 84.48 ± 0.76

68.80 ± 0.43 / 21.84 ± 0.45

81.71 ± 1.12 / 86.29 ± 0.83

89.91 ± 1.81 / 92.38 ± 1.34

12.10.0

12.10.2

12.10.0

12.10.2

meta-llama/Llama-3.1-70B-Instruct (few-shot)

70554

128

131072

True

1,409 ± 457 / 186 ± 63

1.29

83.16 ± 0.92 / 78.18 ± 0.71

69.96 ± 0.89 / 72.99 ± 0.72

50.83 ± 1.22 / 74.22 ± 0.70

60.82 ± 1.25 / 81.99 ± 0.50

69.21 ± 0.07 / 23.55 ± 0.30

77.51 ± 0.75 / 83.07 ± 0.56

83.85 ± 0.63 / 87.76 ± 0.49

14.0.3

Qwen/Qwen2.5-72B-Instruct (few-shot)

72706

152

32768

True

1,219 ± 412 / 158 ± 53

1.33

75.84 ± 1.19 / 69.70 ± 0.91

68.66 ± 1.08 / 74.39 ± 0.68

56.46 ± 2.43 / 76.75 ± 1.60

58.39 ± 1.84 / 81.86 ± 0.68

68.32 ± 0.06 / 20.52 ± 0.14

78.29 ± 1.05 / 83.70 ± 0.78

86.39 ± 0.74 / 89.77 ± 0.56

14.0.4

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

unknown

200

128127

True

916 ± 329 / 114 ± 38

1.33

83.48 ± 1.52 / 75.51 ± 2.26

62.74 ± 2.04 / 71.28 ± 1.46

46.56 ± 3.35 / 71.13 ± 1.97

65.41 ± 1.96 / 84.78 ± 0.68

67.64 ± 0.97 / 20.90 ± 0.92

78.55 ± 1.82 / 83.91 ± 1.34

91.34 ± 1.66 / 93.48 ± 1.27

12.10.0

12.10.2

12.10.0

12.10.2

meta-llama/Meta-Llama-3-70B (few-shot, val)

70554

128

8192

True

312 ± 55 / 177 ± 51

1.35

79.06 ± 2.34 / 74.41 ± 2.24

65.53 ± 2.61 / 68.71 ± 2.08

46.28 ± 4.01 / 72.38 ± 2.33

75.20 ± 1.68 / 86.90 ± 1.14

69.10 ± 0.33 / 21.34 ± 0.43

72.64 ± 1.98 / 79.49 ± 1.48

77.49 ± 4.78 / 82.50 ± 3.93

12.7.0

meta-llama/Llama-3.3-70B-Instruct (few-shot)

70554

128

131072

True

1,353 ± 443 / 180 ± 61

1.38

82.35 ± 0.76 / 76.42 ± 0.70

71.07 ± 0.91 / 72.62 ± 0.74

51.27 ± 1.18 / 74.25 ± 0.77

50.23 ± 1.33 / 76.00 ± 0.67

68.77 ± 0.05 / 22.03 ± 0.24

76.73 ± 0.84 / 82.57 ± 0.63

86.49 ± 0.98 / 89.79 ± 0.76

14.0.3

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

unknown

200

8191

True

637 ± 306 / 92 ± 31

1.41

81.23 ± 1.31 / 72.72 ± 1.69

63.46 ± 1.00 / 69.35 ± 1.28

46.45 ± 4.58 / 72.89 ± 2.30

57.64 ± 1.26 / 81.53 ± 0.66

68.71 ± 0.14 / 20.56 ± 0.22

80.09 ± 2.42 / 85.12 ± 1.78

86.91 ± 1.20 / 90.04 ± 0.92

14.0.3

upstage/SOLAR-10.7B-v1.0 (few-shot)

10732

32

4096

True

3,780 ± 906 / 799 ± 261

1.42

69.50 ± 1.14 / 63.41 ± 0.77

70.01 ± 0.93 / 59.77 ± 0.59

41.35 ± 2.01 / 70.11 ± 0.82

76.79 ± 0.43 / 89.35 ± 0.19

71.16 ± 0.12 / 29.84 ± 0.23

51.95 ± 0.84 / 63.78 ± 0.62

90.94 ± 0.54 / 93.17 ± 0.41

12.5.3

google/gemma-2-27b-it (few-shot)

27227

256

8448

True

1,516 ± 257 / 480 ± 148

1.44

62.40 ± 1.15 / 59.91 ± 1.66

68.68 ± 0.87 / 73.25 ± 0.48

54.17 ± 2.59 / 76.02 ± 1.67

66.96 ± 1.51 / 84.76 ± 0.92

69.79 ± 0.37 / 26.34 ± 0.54

68.71 ± 0.72 / 76.45 ± 0.55

81.25 ± 1.38 / 85.84 ± 1.05

13.0.0

ThatsGroes/gemma-2-27b-it-FP8-Dynamic (few-shot)

28411

256

4096

True

3,633 ± 1,236 / 777 ± 220

1.45

81.06 ± 0.87 / 75.22 ± 1.31

68.92 ± 1.04 / 72.90 ± 0.74

49.06 ± 1.46 / 72.52 ± 0.82

61.27 ± 1.63 / 81.48 ± 0.86

68.00 ± 0.09 / 22.45 ± 0.14

67.84 ± 0.73 / 75.85 ± 0.56

78.26 ± 0.99 / 83.56 ± 0.79

14.0.4

VAGOsolutions/SauerkrautLM-7b-LaserChat (few-shot)

7242

32

4096

False

4,413 ± 1,265 / 551 ± 184

1.53

72.31 ± 0.95 / 58.34 ± 1.84

69.64 ± 0.81 / 74.44 ± 0.54

46.08 ± 1.45 / 71.39 ± 0.74

72.23 ± 1.24 / 86.80 ± 0.57

70.23 ± 0.09 / 26.49 ± 0.30

48.37 ± 1.21 / 61.11 ± 0.87

76.87 ± 0.79 / 82.49 ± 0.62

14.0.4

meta-llama/Llama-2-70b-hf (few-shot, val)

68977

32

4349

True

1,892 ± 650 / 318 ± 105

1.53

72.38 ± 1.28 / 66.44 ± 1.74

60.98 ± 2.19 / 68.64 ± 2.17

43.12 ± 5.21 / 70.12 ± 2.87

74.50 ± 3.41 / 86.22 ± 1.70

71.63 ± 0.35 / 26.83 ± 0.78

54.29 ± 1.91 / 65.78 ± 1.33

74.35 ± 4.59 / 80.47 ± 3.64

12.7.0

meta-llama/Meta-Llama-3-70B-Instruct (few-shot, val)

70554

128

8192

True

1,673 ± 583 / 275 ± 85

1.53

81.30 ± 1.35 / 76.64 ± 1.77

66.18 ± 2.04 / 72.85 ± 1.27

38.10 ± 2.32 / 68.58 ± 1.40

53.31 ± 3.61 / 77.76 ± 1.59

70.35 ± 0.46 / 26.71 ± 0.65

72.39 ± 1.41 / 79.30 ± 1.05

83.86 ± 2.24 / 87.81 ± 1.71

12.7.0

152334H/miqu-1-70b-sf (few-shot, val)

68977

32

33017

True

2,126 ± 676 / 319 ± 104

1.56

71.83 ± 1.48 / 65.37 ± 1.62

62.99 ± 3.04 / 69.81 ± 1.86

39.97 ± 3.89 / 69.38 ± 1.87

64.42 ± 3.17 / 80.97 ± 1.54

71.27 ± 0.50 / 25.32 ± 0.64

64.27 ± 3.04 / 73.20 ± 2.27

77.60 ± 4.02 / 82.85 ± 3.10

12.7.0

VAGOsolutions/FC-SauerkrautLM-7b-beta (few-shot)

unknown

32

8192

False

2,160 ± 514 / 668 ± 256

1.57

70.15 ± 0.85 / 60.29 ± 2.28

69.36 ± 0.87 / 67.37 ± 0.89

37.09 ± 2.04 / 68.05 ± 1.30

77.39 ± 1.14 / 89.74 ± 0.41

69.72 ± 0.17 / 25.82 ± 0.35

50.10 ± 1.05 / 62.63 ± 0.79

77.27 ± 0.64 / 82.91 ± 0.49

12.6.1

google/gemma-2-9b-it (few-shot)

9242

256

8448

True

2,062 ± 397 / 589 ± 178

1.57

58.07 ± 2.46 / 51.39 ± 2.35

68.40 ± 1.17 / 71.79 ± 0.65

51.58 ± 2.03 / 74.57 ± 1.48

62.03 ± 1.62 / 81.91 ± 0.76

69.72 ± 0.42 / 25.66 ± 0.38

63.99 ± 0.76 / 72.84 ± 0.57

75.87 ± 1.74 / 81.61 ± 1.30

13.0.0

CohereForAI/c4ai-command-r-08-2024 (few-shot)

32296

256

131072

False

1,909 ± 646 / 248 ± 84

1.58

78.35 ± 1.42 / 68.56 ± 1.40

67.62 ± 0.95 / 70.27 ± 1.06

46.50 ± 1.03 / 71.67 ± 0.79

63.20 ± 1.15 / 80.12 ± 0.66

70.22 ± 0.23 / 26.49 ± 0.34

60.40 ± 0.83 / 70.22 ± 0.62

60.49 ± 2.10 / 69.11 ± 1.87

14.0.4

gpt-4o-mini-2024-07-18 (few-shot, val)

unknown

200

8191

True

784 ± 310 / 95 ± 28

1.58

77.38 ± 1.61 / 66.97 ± 2.42

66.75 ± 2.50 / 73.93 ± 1.51

52.43 ± 3.69 / 74.46 ± 2.17

41.03 ± 6.90 / 67.96 ± 4.99

68.73 ± 0.21 / 20.97 ± 0.36

66.63 ± 2.82 / 74.92 ± 2.08

79.95 ± 1.44 / 84.69 ± 1.09

14.0.0

14.0.1

mistralai/Mistral-Nemo-Instruct-2407 (few-shot)

12248

131

1024256

True

7,095 ± 2,193 / 1,063 ± 344

1.58

75.24 ± 0.84 / 67.78 ± 2.01

63.05 ± 1.41 / 70.34 ± 0.52

44.75 ± 2.73 / 71.43 ± 1.71

67.70 ± 1.57 / 83.88 ± 1.27

70.56 ± 0.58 / 25.57 ± 0.59

57.55 ± 1.00 / 68.02 ± 0.70

67.97 ± 1.89 / 75.51 ± 1.48

13.0.0

gpt-4o-mini-2024-07-18 (zero-shot, val)

unknown

200

8191

True

908 ± 303 / 96 ± 36

1.61

75.80 ± 0.83 / 52.95 ± 1.04

61.65 ± 2.17 / 71.46 ± 1.72

47.74 ± 4.29 / 73.28 ± 2.23

56.98 ± 1.46 / 79.90 ± 0.76

68.60 ± 0.16 / 20.57 ± 0.28

63.18 ± 2.47 / 72.23 ± 1.82

77.31 ± 1.71 / 82.54 ± 1.37

14.0.1

14.0.3

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

7242

32

4096

False

4,136 ± 1,282 / 668 ± 326

1.65

72.77 ± 1.02 / 57.29 ± 1.58

70.12 ± 0.78 / 74.54 ± 0.50

44.68 ± 0.97 / 71.05 ± 0.52

57.17 ± 2.60 / 80.36 ± 1.40

69.32 ± 0.08 / 23.63 ± 0.24

47.17 ± 1.25 / 60.21 ± 0.92

77.25 ± 0.73 / 82.76 ± 0.56

14.0.4

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

8020

131

32768

True

1,302 ± 323 / 253 ± 86

1.65

72.40 ± 0.80 / 65.83 ± 1.64

63.46 ± 2.10 / 69.49 ± 1.15

35.86 ± 7.94 / 65.20 ± 6.98

68.42 ± 1.21 / 83.97 ± 0.74

69.38 ± 0.48 / 24.96 ± 0.47

53.42 ± 0.91 / 64.98 ± 0.74

78.36 ± 2.31 / 83.62 ± 1.78

13.0.0

skole-gpt-mixtral (few-shot)

unknown

32

32768

False

3,583 ± 977 / 686 ± 231

1.65

67.43 ± 0.90 / 60.40 ± 1.08

68.55 ± 1.35 / 69.63 ± 0.98

39.75 ± 2.28 / 69.19 ± 1.17

65.93 ± 2.77 / 82.86 ± 1.53

70.90 ± 0.34 / 25.88 ± 0.75

59.14 ± 1.02 / 69.28 ± 0.76

57.88 ± 2.80 / 67.75 ± 2.34

13.0.0

gpt-3.5-turbo-0613 (few-shot, val)

unknown

100

4095

True

921 ± 293 / 113 ± 37

1.68

71.48 ± 2.47 / 69.71 ± 1.59

66.41 ± 2.66 / 68.72 ± 1.87

41.43 ± 2.57 / 70.34 ± 1.35

67.90 ± 1.61 / 85.57 ± 0.84

69.57 ± 0.18 / 24.41 ± 0.39

43.69 ± 3.59 / 57.38 ± 3.06

75.60 ± 3.04 / 81.48 ± 2.31

0.0.0

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

unknown

100

8191

True

436 ± 152 / 57 ± 21

1.68

42.40 ± 2.22 / 34.09 ± 2.02

65.24 ± 2.14 / 72.51 ± 1.77

44.59 ± 4.02 / 71.93 ± 2.06

62.94 ± 1.37 / 82.84 ± 0.53

68.33 ± 0.17 / 20.27 ± 0.29

78.80 ± 2.08 / 84.14 ± 1.57

82.60 ± 1.54 / 86.60 ± 1.25

14.0.3

google/gemma-2-9b (few-shot)

9242

256

8448

True

2,038 ± 406 / 566 ± 172

1.71

50.90 ± 2.39 / 44.74 ± 1.46

68.91 ± 0.89 / 70.46 ± 1.23

43.79 ± 3.15 / 71.10 ± 1.74

69.17 ± 1.55 / 84.25 ± 1.13

69.67 ± 0.46 / 25.26 ± 0.45

61.34 ± 0.86 / 70.95 ± 0.66

62.23 ± 3.55 / 70.69 ± 3.11

13.0.0

nvidia/Llama-3.1-Nemotron-70B-Instruct-HF (few-shot)

70554

128

131072

True

1,208 ± 412 / 156 ± 53

1.72

73.66 ± 1.39 / 60.87 ± 1.81

68.56 ± 1.01 / 74.22 ± 0.73

51.33 ± 1.21 / 74.14 ± 0.75

56.87 ± 1.12 / 79.71 ± 0.44

54.04 ± 0.37 / 6.64 ± 0.20

75.50 ± 0.90 / 81.59 ± 0.67

84.61 ± 1.36 / 88.37 ± 1.07

14.0.4

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

8028

256

8192

False

2,686 ± 685 / 491 ± 164

1.75

67.33 ± 1.57 / 53.00 ± 0.88

68.67 ± 0.74 / 66.23 ± 0.49

31.18 ± 1.63 / 65.23 ± 0.69

68.33 ± 2.04 / 84.26 ± 1.04

72.23 ± 0.14 / 30.67 ± 0.16

49.41 ± 1.09 / 62.03 ± 0.82

58.93 ± 1.50 / 68.81 ± 1.21

14.0.4

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

8171

49

4096

True

1,118 ± 302 / 184 ± 63

1.76

66.17 ± 1.74 / 57.76 ± 1.30

68.03 ± 0.81 / 69.65 ± 1.18

39.76 ± 2.51 / 69.54 ± 1.18

71.21 ± 1.53 / 84.07 ± 0.91

69.30 ± 0.10 / 23.68 ± 0.29

49.99 ± 0.60 / 62.20 ± 0.47

52.45 ± 1.74 / 61.79 ± 1.58

14.0.4

mistralai/Mixtral-8x7B-Instruct-v0.1 (few-shot)

46703

32

32768

True

5,535 ± 1,837 / 760 ± 256

1.76

70.02 ± 1.35 / 60.95 ± 1.57

69.48 ± 0.68 / 71.36 ± 1.02

44.59 ± 0.99 / 70.37 ± 0.77

55.70 ± 2.73 / 77.76 ± 1.18

68.79 ± 0.10 / 21.92 ± 0.16

55.61 ± 0.89 / 66.57 ± 0.68

57.52 ± 1.56 / 67.66 ± 1.21

14.0.4

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

8030

128

131328

True

1,473 ± 377 / 283 ± 96

1.77

76.95 ± 0.95 / 72.47 ± 0.82

68.12 ± 0.92 / 72.48 ± 0.53

34.34 ± 3.37 / 65.84 ± 1.59

47.88 ± 3.37 / 76.21 ± 1.69

69.57 ± 0.25 / 26.30 ± 0.35

56.62 ± 0.49 / 67.29 ± 0.39

69.03 ± 1.19 / 76.69 ± 0.89

13.0.0

nvidia/mistral-nemo-minitron-8b-base (few-shot)

8414

131

8192

True

2,470 ± 836 / 326 ± 111

1.77

67.52 ± 3.88 / 61.41 ± 2.66

69.03 ± 0.69 / 65.72 ± 2.41

40.51 ± 1.93 / 69.82 ± 0.89

58.12 ± 2.97 / 72.32 ± 2.85

67.86 ± 0.55 / 19.31 ± 0.44

55.93 ± 0.71 / 66.85 ± 0.53

64.11 ± 1.73 / 72.26 ± 1.43

14.1.1

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

6061

64

4352

True

2,867 ± 550 / 793 ± 253

1.78

55.51 ± 1.48 / 50.61 ± 1.37

68.74 ± 0.81 / 58.25 ± 0.31

40.81 ± 3.66 / 69.94 ± 1.98

72.33 ± 2.68 / 84.50 ± 1.61

67.71 ± 0.08 / 23.57 ± 0.14

48.36 ± 0.95 / 61.21 ± 0.76

68.50 ± 0.63 / 76.04 ± 0.47

13.0.0

meta-llama/Llama-2-70b-chat-hf (few-shot, val)

68977

32

4096

True

1,979 ± 621 / 320 ± 105

1.81

72.80 ± 1.65 / 64.88 ± 3.12

63.76 ± 2.53 / 71.14 ± 1.90

28.37 ± 4.76 / 62.85 ± 3.04

64.70 ± 2.69 / 81.80 ± 1.41

71.04 ± 0.42 / 25.68 ± 0.52

47.00 ± 2.21 / 60.12 ± 1.58

61.56 ± 2.97 / 70.27 ± 2.27

12.7.0

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

8030

128

8192

True

1,483 ± 377 / 287 ± 97

1.82

75.02 ± 1.31 / 69.47 ± 1.18

67.64 ± 1.12 / 71.04 ± 1.17

32.29 ± 3.05 / 64.85 ± 2.07

54.84 ± 2.22 / 79.10 ± 1.10

69.28 ± 0.17 / 25.48 ± 0.61

53.77 ± 1.03 / 64.91 ± 0.80

57.64 ± 1.55 / 67.29 ± 1.30

12.6.1

NorwAI/NorwAI-Mixtral-8x7B-instruct (few-shot)

46998

68

32768

True

9,015 ± 2,966 / 1,121 ± 510

1.83

63.77 ± 2.22 / 57.86 ± 1.49

69.23 ± 0.98 / 66.61 ± 1.42

38.49 ± 4.39 / 65.15 ± 4.44

57.03 ± 4.40 / 76.18 ± 2.59

70.47 ± 0.55 / 26.63 ± 0.50

54.58 ± 0.94 / 65.69 ± 0.71

47.93 ± 3.99 / 59.68 ± 3.43

14.0.4

mistralai/Mixtral-8x7B-v0.1 (few-shot)

46703

32

32768

True

2,363 ± 794 / 311 ± 105

1.83

65.31 ± 3.21 / 60.19 ± 2.58

68.87 ± 0.60 / 66.34 ± 0.90

43.07 ± 1.73 / 70.80 ± 1.06

63.97 ± 3.33 / 75.92 ± 3.61

68.68 ± 1.07 / 22.56 ± 1.09

56.64 ± 0.82 / 67.35 ± 0.65

33.42 ± 5.53 / 47.54 ± 4.76

14.0.4

robinsmits/Qwen1.5-7B-Dutch-Chat-Sft-Bf16 (few-shot)

7719

152

32768

False

2,413 ± 463 / 700 ± 220

1.85

67.52 ± 1.19 / 59.09 ± 2.64

63.10 ± 1.92 / 70.11 ± 0.80

37.75 ± 2.52 / 67.53 ± 1.75

64.88 ± 1.60 / 82.23 ± 0.72

68.45 ± 0.22 / 21.67 ± 0.52

47.66 ± 0.96 / 60.75 ± 0.72

63.35 ± 1.62 / 72.16 ± 1.31

12.6.1

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

8030

128

131328

True

2,986 ± 823 / 276 ± 94

1.87

69.86 ± 2.10 / 62.68 ± 2.21

66.76 ± 0.72 / 68.58 ± 0.72

30.96 ± 2.46 / 61.29 ± 3.61

71.39 ± 2.20 / 84.24 ± 1.55

67.93 ± 0.44 / 22.00 ± 0.51

52.47 ± 0.85 / 64.25 ± 0.66

43.95 ± 3.26 / 57.04 ± 2.74

12.11.0

13.0.0

CohereForAI/c4ai-command-r-v01 (few-shot)

34981

256

8192

False

1,919 ± 645 / 248 ± 83

1.88

71.96 ± 0.83 / 63.08 ± 1.18

67.26 ± 0.99 / 68.90 ± 1.29

40.81 ± 1.97 / 70.25 ± 1.02

49.79 ± 2.48 / 73.04 ± 1.31

68.97 ± 0.15 / 22.78 ± 0.28

50.59 ± 1.17 / 62.68 ± 0.84

58.04 ± 1.68 / 68.27 ± 1.28

14.0.4

14.1.1

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

8030

128

131072

True

2,996 ± 817 / 284 ± 96

1.89

59.28 ± 0.97 / 55.81 ± 1.40

68.49 ± 0.77 / 58.77 ± 0.44

21.95 ± 3.10 / 49.75 ± 3.85

78.16 ± 2.12 / 87.44 ± 1.32

69.54 ± 0.30 / 26.96 ± 0.56

46.35 ± 0.95 / 59.11 ± 0.72

60.61 ± 4.03 / 67.80 ± 3.65

13.2.0

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

7242

32

32768

False

5,993 ± 1,028 / 1,742 ± 561

1.89

70.62 ± 0.90 / 58.92 ± 2.15

67.78 ± 1.03 / 72.29 ± 0.47

30.99 ± 2.94 / 62.20 ± 2.56

49.56 ± 2.85 / 76.72 ± 1.15

70.76 ± 0.56 / 24.95 ± 1.03

44.11 ± 4.39 / 57.76 ± 3.36

69.20 ± 2.71 / 76.75 ± 2.06

12.6.1

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

3821

32

131072

True

7,312 ± 1,668 / 1,609 ± 525

1.92

64.09 ± 0.96 / 49.92 ± 2.47

46.77 ± 4.36 / 60.99 ± 2.15

31.62 ± 2.25 / 63.73 ± 1.79

71.25 ± 0.83 / 85.72 ± 0.57

69.54 ± 0.57 / 24.79 ± 0.89

57.66 ± 1.17 / 68.22 ± 0.88

72.26 ± 0.86 / 79.10 ± 0.64

12.9.1

12.10.0

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

8030

128

8192

True

1,477 ± 376 / 285 ± 97

1.95

66.31 ± 2.09 / 58.68 ± 1.95

64.30 ± 0.65 / 69.26 ± 0.50

28.18 ± 3.96 / 58.97 ± 4.03

70.38 ± 3.51 / 82.95 ± 2.38

67.90 ± 0.49 / 21.54 ± 0.57

52.54 ± 0.88 / 64.26 ± 0.66

41.19 ± 4.40 / 54.78 ± 3.62

12.6.1

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

7242

32

32768

True

1,446 ± 354 / 295 ± 100

1.95

63.40 ± 2.72 / 56.92 ± 2.17

68.17 ± 1.33 / 70.74 ± 0.93

30.92 ± 4.81 / 63.79 ± 4.42

73.45 ± 1.83 / 84.54 ± 1.42

69.11 ± 0.31 / 23.70 ± 0.53

47.74 ± 1.26 / 60.63 ± 0.97

34.96 ± 4.19 / 49.62 ± 3.28

9.1.2

12.5.1

11.0.0

9.2.0

alpindale/Mistral-7B-v0.2-hf (few-shot)

7242

32

32768

True

1,841 ± 297 / 651 ± 193

1.96

61.02 ± 2.70 / 55.57 ± 2.50

67.10 ± 0.81 / 70.66 ± 0.76

29.82 ± 5.18 / 62.86 ± 4.72

73.50 ± 1.67 / 84.26 ± 1.30

69.02 ± 0.55 / 23.87 ± 0.72

47.13 ± 1.24 / 60.08 ± 0.92

35.88 ± 4.42 / 49.95 ± 3.87

12.5.2

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

7248

33

32768

True

1,364 ± 343 / 266 ± 90

1.96

61.02 ± 2.74 / 55.65 ± 2.55

67.29 ± 0.80 / 70.81 ± 0.84

30.10 ± 5.12 / 62.99 ± 4.71

73.59 ± 1.75 / 84.31 ± 1.35

69.04 ± 0.56 / 23.86 ± 0.77

47.63 ± 1.17 / 60.45 ± 0.88

35.63 ± 4.21 / 49.72 ± 3.73

12.10.4

12.10.5

12.10.4

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

2534

49

4352

True

10,187 ± 2,363 / 2,204 ± 737

1.97

57.78 ± 1.91 / 50.30 ± 2.29

67.81 ± 1.00 / 61.89 ± 1.51

22.81 ± 2.71 / 60.07 ± 2.49

72.90 ± 1.16 / 83.67 ± 0.66

67.33 ± 0.41 / 22.76 ± 0.60

37.80 ± 0.99 / 53.52 ± 0.78

67.23 ± 0.96 / 75.34 ± 0.74

13.0.0

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

8028

256

8192

False

2,707 ± 688 / 497 ± 166

1.98

56.16 ± 3.59 / 51.12 ± 2.58

68.27 ± 0.53 / 60.37 ± 0.57

23.82 ± 2.16 / 60.83 ± 1.63

74.23 ± 1.31 / 85.11 ± 0.73

72.14 ± 0.13 / 29.83 ± 0.23

33.04 ± 1.30 / 49.51 ± 0.98

55.10 ± 1.39 / 64.78 ± 1.34

14.0.4

14.1.2

14.0.4

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

8171

49

4096

True

2,515 ± 625 / 476 ± 159

1.98

55.76 ± 2.15 / 52.69 ± 1.24

66.89 ± 1.11 / 69.52 ± 0.94

36.60 ± 2.37 / 67.85 ± 1.19

67.55 ± 1.46 / 80.25 ± 1.22

66.51 ± 0.93 / 21.81 ± 1.31

46.22 ± 1.20 / 59.52 ± 0.94

52.23 ± 2.04 / 61.87 ± 1.91

14.0.4

14.1.2

14.0.4

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

2634

49

4352

True

10,194 ± 2,403 / 2,193 ± 731

2.00

61.97 ± 1.74 / 54.58 ± 1.53

67.54 ± 1.33 / 66.16 ± 1.54

31.70 ± 2.00 / 65.43 ± 1.09

59.78 ± 1.66 / 78.74 ± 0.76

68.61 ± 0.41 / 24.36 ± 0.58

39.82 ± 0.91 / 54.59 ± 0.66

57.51 ± 0.92 / 67.67 ± 0.71

13.0.0

occiglot/occiglot-7b-de-en-instruct (few-shot)

7242

32

32768

False

1,584 ± 217 / 635 ± 178

2.00

60.90 ± 3.17 / 53.69 ± 2.67

66.54 ± 1.42 / 69.26 ± 0.83

23.60 ± 3.72 / 57.65 ± 4.08

75.14 ± 1.51 / 87.25 ± 0.61

68.96 ± 0.53 / 25.60 ± 0.80

42.53 ± 0.85 / 56.25 ± 0.72

41.79 ± 3.18 / 54.82 ± 2.83

12.5.2

12.3.1

12.4.0

12.3.1

mgoin/Nemotron-4-340B-Instruct-hf-FP8 (few-shot)

341029

256

4096

True

1,904 ± 475 / 361 ± 121

2.02

51.92 ± 11.90 / 40.30 ± 8.36

67.01 ± 2.29 / 72.21 ± 1.61

45.98 ± 7.10 / 68.89 ± 6.52

30.12 ± 6.60 / 48.37 ± 9.50

65.82 ± 1.56 / 18.90 ± 1.25

60.71 ± 5.40 / 69.77 ± 4.90

77.14 ± 11.59 / 81.19 ± 10.68

14.0.4

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

3213

128

131328

False

10,424 ± 2,641 / 2,081 ± 666

2.03

68.44 ± 1.14 / 56.80 ± 2.31

66.00 ± 1.41 / 70.47 ± 1.02

32.04 ± 3.44 / 65.62 ± 1.88

49.54 ± 3.13 / 75.26 ± 1.64

69.23 ± 0.19 / 24.84 ± 0.39

45.50 ± 0.74 / 58.96 ± 0.57

46.50 ± 1.78 / 59.43 ± 1.44

13.0.0

google/gemma-7b (few-shot)

8538

256

8320

True

1,378 ± 260 / 387 ± 119

2.04

47.20 ± 3.34 / 41.96 ± 3.12

63.88 ± 2.24 / 66.49 ± 0.81

35.75 ± 2.65 / 66.41 ± 2.26

69.40 ± 4.29 / 82.46 ± 2.86

69.25 ± 0.80 / 24.32 ± 0.95

50.57 ± 1.00 / 61.48 ± 0.80

35.79 ± 4.89 / 46.37 ± 3.67

12.9.1

12.10.0

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

3950

152

32768

True

3,248 ± 739 / 761 ± 252

2.06

55.45 ± 1.61 / 49.72 ± 1.27

60.55 ± 1.54 / 68.53 ± 0.71

28.60 ± 3.36 / 60.31 ± 4.06

70.49 ± 0.74 / 82.68 ± 0.52

68.67 ± 0.11 / 21.65 ± 0.21

39.82 ± 0.76 / 54.62 ± 0.57

51.82 ± 1.03 / 63.79 ± 0.79

12.5.2

10.0.1

12.1.0

claude-3-5-sonnet-20241022 (zero-shot, val)

unknown

200000

True

193 ± 87 / 55 ± 19

2.06

82.11 ± 1.42 / 79.77 ± 1.03

67.01 ± 2.34 / 71.49 ± 1.89

51.09 ± 5.20 / 74.48 ± 2.70

52.41 ± 1.56 / 77.79 ± 1.01

67.73 ± 0.14 / 18.28 ± 0.26

26.14 ± 2.01 / 35.16 ± 1.54

13.56 ± 3.11 / 26.48 ± 1.71

14.0.3

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

7242

32

32768

False

634 ± 179 / 110 ± 35

2.10

57.58 ± 2.30 / 47.94 ± 2.89

61.44 ± 2.02 / 69.47 ± 0.98

34.92 ± 2.40 / 66.67 ± 1.41

65.38 ± 1.76 / 81.90 ± 0.57

69.62 ± 0.31 / 24.65 ± 0.44

38.40 ± 0.98 / 53.43 ± 0.76

35.72 ± 1.56 / 49.69 ± 1.42

9.3.1

12.4.0

9.3.1

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

7242

32

32768

True

5,054 ± 1,200 / 1,056 ± 339

2.10

59.10 ± 1.87 / 51.31 ± 1.87

68.41 ± 1.17 / 70.85 ± 0.74

25.43 ± 4.22 / 60.79 ± 3.45

71.89 ± 2.20 / 82.99 ± 1.78

67.99 ± 0.41 / 22.12 ± 0.52

44.09 ± 1.21 / 56.37 ± 0.96

32.29 ± 4.57 / 45.16 ± 4.28

12.5.3

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

9152

128

4096

False

1,483 ± 321 / 379 ± 158

2.10

44.81 ± 2.07 / 39.51 ± 1.62

62.54 ± 2.18 / 68.78 ± 1.18

28.10 ± 4.55 / 62.46 ± 4.36

71.71 ± 2.21 / 84.31 ± 1.69

69.27 ± 0.77 / 25.38 ± 1.10

45.95 ± 0.80 / 59.04 ± 0.60

43.97 ± 1.97 / 56.43 ± 1.61

13.1.0

RuterNorway/Llama-2-13b-chat-norwegian (few-shot)

unknown

32

4096

False

3,254 ± 1,068 / 484 ± 173

2.11

64.93 ± 2.24 / 57.95 ± 1.24

64.14 ± 1.61 / 68.00 ± 1.67

28.08 ± 3.86 / 62.71 ± 2.98

62.09 ± 1.68 / 79.57 ± 0.96

68.84 ± 0.52 / 24.99 ± 0.80

36.49 ± 0.82 / 52.02 ± 0.62

38.09 ± 2.61 / 52.60 ± 2.19

9.3.1

occiglot/occiglot-7b-de-en (few-shot)

7242

32

32768

True

1,992 ± 319 / 706 ± 211

2.11

56.07 ± 4.01 / 51.66 ± 3.99

65.29 ± 1.43 / 69.38 ± 0.73

25.78 ± 2.57 / 61.50 ± 2.24

73.13 ± 4.05 / 83.85 ± 2.69

68.76 ± 0.54 / 23.14 ± 0.51

41.47 ± 0.94 / 55.81 ± 0.77

32.75 ± 3.47 / 48.17 ± 2.83

12.3.2

12.3.1

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

3950

152

32768

False

4,347 ± 893 / 1,135 ± 365

2.13

58.56 ± 1.99 / 54.14 ± 2.01

59.62 ± 1.29 / 68.55 ± 0.56

28.55 ± 3.97 / 60.49 ± 4.25

70.04 ± 0.89 / 82.09 ± 0.84

67.27 ± 0.39 / 19.77 ± 0.31

38.68 ± 1.02 / 53.45 ± 0.74

47.47 ± 1.28 / 60.44 ± 1.02

12.5.2

10.0.1

12.1.0

12.5.2

12.1.0

meta-llama/Llama-2-13b-chat-hf (few-shot)

13016

32

4320

True

2,849 ± 622 / 723 ± 229

2.13

68.75 ± 1.54 / 60.01 ± 1.57

62.37 ± 1.68 / 67.79 ± 1.51

25.07 ± 3.20 / 61.02 ± 2.86

61.56 ± 1.18 / 80.11 ± 0.70

69.40 ± 0.49 / 24.79 ± 0.61

38.00 ± 0.83 / 52.95 ± 0.66

42.14 ± 2.47 / 55.96 ± 1.93

12.11.0

12.10.4

12.11.0

12.10.4

meta-llama/Llama-2-13b-hf (few-shot)

13016

32

4096

True

2,898 ± 637 / 736 ± 236

2.16

63.75 ± 2.50 / 59.28 ± 2.20

61.85 ± 1.70 / 69.02 ± 0.76

26.41 ± 4.93 / 59.67 ± 4.84

73.48 ± 2.06 / 85.27 ± 1.15

67.73 ± 0.58 / 22.39 ± 0.65

38.04 ± 1.30 / 53.30 ± 0.99

28.16 ± 3.07 / 44.80 ± 2.46

12.10.5

12.10.4

12.10.5

12.10.4

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

7242

32

32768

False

2,370 ± 416 / 711 ± 242

2.17

62.11 ± 1.61 / 52.36 ± 2.00

59.91 ± 2.10 / 68.92 ± 1.21

30.66 ± 3.60 / 64.32 ± 2.03

58.27 ± 2.09 / 77.85 ± 0.70

69.75 ± 0.63 / 24.71 ± 0.72

34.93 ± 1.35 / 50.71 ± 1.00

44.91 ± 2.44 / 58.07 ± 1.93

9.3.1

9.2.0

9.3.1

12.4.0

9.3.2

google/gemma-7b-it (few-shot)

8538

256

8445

False

1,792 ± 249 / 668 ± 203

2.18

66.70 ± 0.99 / 61.08 ± 1.16

55.62 ± 2.54 / 64.98 ± 2.03

31.36 ± 2.63 / 65.21 ± 1.16

72.58 ± 0.68 / 84.67 ± 0.91

67.24 ± 0.71 / 21.99 ± 0.99

35.27 ± 1.22 / 50.12 ± 0.94

32.54 ± 1.77 / 46.58 ± 1.66

12.10.0

nvidia/mistral-nemo-minitron-8b-instruct (few-shot)

8414

131

8192

True

3,161 ± 676 / 1,247 ± 481

2.18

72.63 ± 1.76 / 63.75 ± 1.29

65.74 ± 1.14 / 69.65 ± 0.96

43.43 ± 1.62 / 69.19 ± 1.28

0.00 ± 0.00 / 20.81 ± 0.82

66.76 ± 0.07 / 17.17 ± 0.16

60.00 ± 0.90 / 69.95 ± 0.66

69.36 ± 1.38 / 76.63 ± 1.08

14.1.1

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

7242

32

32768

False

2,088 ± 352 / 706 ± 214

2.18

56.90 ± 3.08 / 51.16 ± 2.56

62.10 ± 1.65 / 68.81 ± 0.76

20.17 ± 3.68 / 54.76 ± 4.24

75.29 ± 1.37 / 86.48 ± 0.72

69.63 ± 0.46 / 25.61 ± 0.45

38.48 ± 1.10 / 53.32 ± 0.83

27.67 ± 2.08 / 43.61 ± 2.08

12.5.2

12.2.0

12.3.1

12.4.0

12.3.1

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

3821

32

2047

True

8,681 ± 1,650 / 2,177 ± 717

2.19

68.69 ± 1.08 / 58.53 ± 1.82

66.77 ± 0.54 / 70.41 ± 0.88

42.14 ± 2.29 / 70.53 ± 0.99

63.71 ± 1.70 / 81.26 ± 1.09

45.05 ± 10.76 / 15.91 ± 3.64

56.92 ± 0.88 / 67.56 ± 0.66

55.78 ± 0.98 / 65.88 ± 0.81

14.0.4

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

2614

256

8448

True

5,374 ± 1,233 / 1,193 ± 377

2.20

44.36 ± 1.46 / 38.51 ± 1.53

66.37 ± 1.35 / 67.64 ± 1.35

34.69 ± 2.50 / 67.16 ± 1.29

55.07 ± 1.73 / 78.88 ± 0.80

68.09 ± 0.20 / 23.58 ± 0.33

42.89 ± 1.25 / 57.03 ± 0.94

45.52 ± 0.68 / 58.39 ± 0.55

13.0.0

danish-foundation-models/munin-7b-v0.1dev0 (few-shot)

7242

32

8192

True

6,113 ± 1,044 / 1,790 ± 579

2.22

56.38 ± 2.95 / 50.80 ± 2.82

66.04 ± 1.68 / 65.21 ± 1.48

22.15 ± 3.57 / 57.71 ± 4.24

71.32 ± 1.41 / 83.70 ± 0.93

68.13 ± 0.68 / 23.08 ± 0.59

35.47 ± 1.23 / 50.47 ± 0.94

29.25 ± 3.05 / 44.01 ± 2.75

12.5.2

12.5.0

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

6738

32

4096

False

2,643 ± 455 / 800 ± 247

2.28

62.53 ± 1.35 / 53.42 ± 2.04

62.23 ± 1.29 / 68.09 ± 1.34

22.71 ± 1.81 / 60.79 ± 1.08

64.45 ± 1.39 / 80.79 ± 0.79

69.95 ± 0.32 / 25.22 ± 0.33

30.47 ± 0.70 / 46.82 ± 0.55

30.18 ± 1.87 / 45.85 ± 1.98

9.3.1

12.4.0

9.3.1

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

3374

49

4096

True

10,246 ± 3,021 / 1,629 ± 550

2.29

52.79 ± 4.09 / 43.45 ± 2.82

65.92 ± 1.02 / 70.47 ± 0.75

16.74 ± 2.67 / 55.45 ± 3.35

64.92 ± 2.84 / 80.88 ± 1.27

65.50 ± 0.97 / 21.90 ± 0.40

33.84 ± 0.87 / 50.07 ± 0.67

49.84 ± 1.22 / 61.87 ± 1.03

13.0.0

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

8055

49

4096

True

2,313 ± 423 / 682 ± 210

2.30

72.76 ± 0.57 / 67.49 ± 0.90

62.35 ± 1.68 / 67.34 ± 0.77

21.57 ± 2.22 / 59.22 ± 1.69

69.80 ± 3.60 / 81.37 ± 2.26

67.73 ± 0.60 / 23.48 ± 0.91

25.63 ± 0.81 / 44.26 ± 0.63

16.44 ± 1.84 / 36.73 ± 1.54

13.0.0

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

8055

49

4351

True

1,002 ± 95 / 416 ± 105

2.30

72.64 ± 0.61 / 67.38 ± 0.91

62.31 ± 1.79 / 67.31 ± 0.86

22.38 ± 2.05 / 59.67 ± 1.50

69.84 ± 3.57 / 81.40 ± 2.25

67.71 ± 0.62 / 23.43 ± 0.95

25.41 ± 0.75 / 44.10 ± 0.58

16.50 ± 1.79 / 36.76 ± 1.44

12.10.5

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

7242

32

32768

True

2,219 ± 427 / 717 ± 224

2.30

55.37 ± 2.94 / 51.08 ± 2.87

63.32 ± 1.29 / 68.50 ± 0.53

18.92 ± 2.39 / 57.96 ± 1.89

72.38 ± 2.57 / 83.46 ± 1.49

68.61 ± 0.55 / 23.48 ± 0.74

37.04 ± 1.33 / 52.33 ± 1.02

23.54 ± 2.08 / 40.78 ± 1.59

12.5.2

12.1.0

12.2.0

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

6738

32

4096

True

3,136 ± 558 / 942 ± 290

2.32

53.21 ± 1.85 / 45.68 ± 2.60

65.98 ± 1.15 / 64.68 ± 2.08

7.26 ± 2.04 / 52.39 ± 1.92

64.71 ± 1.95 / 79.37 ± 1.16

68.81 ± 0.21 / 24.11 ± 0.23

36.50 ± 1.02 / 52.05 ± 0.77

46.95 ± 1.41 / 59.90 ± 1.18

13.2.0

13.0.0

13.2.0

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

3374

49

4096

True

10,504 ± 3,028 / 1,678 ± 559

2.34

49.44 ± 3.68 / 39.69 ± 2.34

66.65 ± 1.04 / 65.72 ± 1.32

12.56 ± 2.15 / 54.20 ± 3.42

63.29 ± 4.61 / 75.62 ± 3.79

66.38 ± 0.28 / 21.17 ± 0.41

32.06 ± 0.87 / 49.12 ± 0.61

58.21 ± 0.98 / 68.63 ± 0.74

13.0.0

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

8055

49

4096

True

5,617 ± 995 / 1,623 ± 540

2.35

72.59 ± 0.91 / 65.83 ± 1.30

61.61 ± 1.45 / 67.09 ± 1.38

18.37 ± 2.07 / 56.26 ± 2.62

66.68 ± 3.56 / 78.95 ± 2.38

68.41 ± 0.33 / 24.66 ± 0.47

24.14 ± 0.58 / 42.17 ± 0.33

14.42 ± 2.00 / 34.50 ± 1.81

13.0.0

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

3213

128

131328

True

3,713 ± 877 / 836 ± 267

2.35

59.09 ± 1.44 / 52.03 ± 1.96

63.29 ± 1.29 / 67.82 ± 0.74

13.50 ± 4.14 / 50.33 ± 5.61

68.15 ± 2.21 / 81.12 ± 1.18

67.73 ± 0.51 / 22.02 ± 0.74

40.10 ± 0.68 / 54.53 ± 0.48

20.88 ± 2.25 / 39.75 ± 1.62

13.0.0

stabilityai/stablelm-2-1_6b (few-shot)

1645

100

4096

True

7,259 ± 2,120 / 1,240 ± 432

2.35

47.50 ± 1.70 / 41.85 ± 1.68

64.69 ± 1.33 / 57.60 ± 0.63

8.01 ± 1.97 / 51.78 ± 1.65

71.81 ± 1.05 / 80.90 ± 0.72

69.71 ± 0.11 / 27.72 ± 0.26

18.92 ± 0.73 / 38.74 ± 0.53

54.53 ± 0.89 / 65.79 ± 0.65

12.10.8

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

6738

32

4096

True

930 ± 310 / 128 ± 43

2.37

55.27 ± 2.79 / 50.25 ± 2.12

65.16 ± 1.21 / 66.86 ± 1.32

20.43 ± 3.69 / 55.98 ± 4.88

69.82 ± 2.49 / 81.43 ± 1.73

68.82 ± 0.54 / 23.58 ± 0.59

25.98 ± 0.90 / 42.52 ± 0.62

11.77 ± 1.81 / 31.25 ± 1.99

9.2.0

12.5.1

12.0.0

9.2.0

microsoft/phi-2 (few-shot)

2780

51

2048

True

3,472 ± 885 / 728 ± 239

2.38

49.16 ± 3.09 / 43.10 ± 2.58

62.41 ± 1.51 / 67.24 ± 1.18

12.31 ± 2.96 / 48.73 ± 5.08

75.79 ± 1.88 / 86.40 ± 1.10

67.79 ± 0.51 / 23.29 ± 0.89

40.15 ± 1.11 / 54.65 ± 0.86

23.21 ± 2.87 / 39.96 ± 2.45

12.5.2

12.2.0

12.0.0

NorwAI/NorwAI-Mixtral-8x7B (few-shot)

46998

68

32768

True

2,368 ± 793 / 317 ± 108

2.40

63.12 ± 2.34 / 56.38 ± 1.23

66.47 ± 1.61 / 69.35 ± 1.31

38.82 ± 3.65 / 67.96 ± 2.36

29.16 ± 3.45 / 46.82 ± 4.35

57.65 ± 5.62 / 18.79 ± 1.79

54.57 ± 1.21 / 65.11 ± 0.83

30.13 ± 5.52 / 42.55 ± 4.99

14.0.4

allenai/OLMo-1.7-7B-hf (few-shot)

6888

50

4096

True

3,371 ± 876 / 561 ± 184

2.41

41.02 ± 3.86 / 40.03 ± 2.56

66.43 ± 0.78 / 57.25 ± 0.33

5.17 ± 1.55 / 37.03 ± 2.40

76.04 ± 0.97 / 85.31 ± 0.82

70.28 ± 0.06 / 28.85 ± 0.17

34.92 ± 0.75 / 50.82 ± 0.73

18.73 ± 1.32 / 34.76 ± 0.91

12.10.5

12.10.4

12.10.5

12.10.4

Rijgersberg/GEITje-7B (few-shot)

7242

32

32768

True

5,887 ± 1,087 / 1,600 ± 522

2.44

53.39 ± 2.97 / 47.76 ± 2.67

65.21 ± 1.35 / 65.73 ± 1.61

12.63 ± 2.60 / 50.10 ± 3.87

65.74 ± 2.31 / 77.95 ± 1.84

68.05 ± 0.54 / 24.11 ± 0.59

31.65 ± 1.00 / 47.19 ± 0.83

17.69 ± 2.71 / 35.94 ± 2.08

9.3.1

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

1711

49

8192

True

15,971 ± 3,654 / 3,609 ± 1,197

2.49

47.58 ± 2.41 / 39.52 ± 1.41

66.78 ± 0.76 / 61.52 ± 0.99

20.53 ± 3.83 / 58.47 ± 2.90

58.07 ± 2.23 / 69.96 ± 1.63

62.45 ± 5.62 / 19.60 ± 1.85

32.90 ± 0.96 / 49.16 ± 0.95

25.32 ± 2.12 / 42.50 ± 2.39

13.1.0

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

6738

32

2048

True

4,405 ± 1,098 / 1,032 ± 345

2.54

47.76 ± 2.72 / 44.84 ± 2.71

66.41 ± 0.85 / 65.96 ± 2.20

5.76 ± 1.50 / 50.36 ± 2.34

70.34 ± 1.81 / 81.54 ± 1.06

65.88 ± 1.55 / 19.75 ± 1.04

29.83 ± 1.09 / 47.45 ± 0.69

18.44 ± 2.21 / 36.96 ± 2.10

12.10.5

12.10.8

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

6738

32

4096

True

6,275 ± 1,193 / 1,755 ± 578

2.59

47.20 ± 3.39 / 44.00 ± 3.43

64.82 ± 1.13 / 67.11 ± 1.41

7.57 ± 2.31 / 45.83 ± 3.97

73.88 ± 1.02 / 83.56 ± 0.75

67.34 ± 0.38 / 21.61 ± 0.61

16.59 ± 1.01 / 35.35 ± 0.70

11.97 ± 1.35 / 32.72 ± 0.94

13.0.0

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

1837

152

32768

True

5,666 ± 1,328 / 1,256 ± 408

2.59

37.22 ± 3.24 / 34.07 ± 3.11

64.34 ± 1.18 / 62.90 ± 1.36

15.30 ± 1.17 / 55.67 ± 1.16

64.41 ± 1.29 / 75.79 ± 0.97

68.15 ± 0.13 / 22.11 ± 0.19

27.24 ± 1.14 / 44.95 ± 0.84

22.84 ± 1.16 / 41.75 ± 0.99

12.5.2

10.0.1

12.1.0

claude-3-5-haiku-20241022 (zero-shot, val)

unknown

200000

True

277 ± 77 / 70 ± 25

2.60

74.35 ± 1.24 / 63.12 ± 1.09

31.19 ± 2.30 / 44.96 ± 1.92

21.76 ± 3.56 / 59.95 ± 1.92

45.70 ± 1.03 / 74.84 ± 0.76

67.96 ± 0.14 / 18.65 ± 0.23

28.80 ± 2.50 / 39.84 ± 1.87

42.52 ± 3.28 / 53.63 ± 2.58

14.0.3

14.0.2

14.0.3

NbAiLab/nb-llama-3.1-70B (few-shot)

70554

128

131072

True

1,220 ± 411 / 158 ± 53

2.62

76.84 ± 1.89 / 72.90 ± 1.50

67.91 ± 1.19 / 68.81 ± 1.37

30.61 ± 4.10 / 62.57 ± 4.17

0.10 ± 0.04 / 16.48 ± 1.10

58.37 ± 1.69 / 13.47 ± 1.23

48.18 ± 2.66 / 61.07 ± 2.06

34.99 ± 1.90 / 47.61 ± 1.78

14.0.4

NbAiLab/nb-llama-3.1-8B (few-shot)

8030

128

131072

True

1,297 ± 338 / 245 ± 83

2.62

62.90 ± 2.49 / 56.12 ± 1.92

62.26 ± 1.62 / 66.24 ± 0.84

24.37 ± 1.47 / 60.05 ± 1.82

25.93 ± 4.95 / 47.47 ± 4.90

63.31 ± 0.93 / 15.85 ± 0.47

40.90 ± 1.22 / 55.74 ± 0.97

16.59 ± 3.39 / 36.28 ± 2.78

14.0.4

14.1.2

14.0.4

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

3483

49

2048

True

9,059 ± 1,947 / 2,201 ± 728

2.62

58.30 ± 2.16 / 53.45 ± 1.98

59.01 ± 2.14 / 62.37 ± 1.73

10.33 ± 3.06 / 53.28 ± 2.21

65.04 ± 2.20 / 75.93 ± 1.50

67.46 ± 0.31 / 23.04 ± 0.42

14.10 ± 0.93 / 34.24 ± 0.84

10.67 ± 0.81 / 32.66 ± 0.64

13.0.0

HuggingFaceTB/SmolLM2-1.7B (few-shot)

1711

49

8192

True

16,249 ± 3,690 / 3,689 ± 1,226

2.67

41.57 ± 4.29 / 37.51 ± 3.05

62.32 ± 1.12 / 67.09 ± 0.96

8.04 ± 3.17 / 48.16 ± 5.38

56.01 ± 3.10 / 67.43 ± 2.41

65.06 ± 0.87 / 20.12 ± 0.51

34.02 ± 0.87 / 50.12 ± 0.76

22.81 ± 1.86 / 41.37 ± 1.41

13.1.0

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

2614

256

8448

True

5,235 ± 1,226 / 1,154 ± 366

2.67

30.82 ± 4.46 / 27.49 ± 3.87

66.28 ± 0.84 / 65.60 ± 1.23

10.09 ± 2.52 / 48.92 ± 3.34

64.96 ± 3.77 / 78.21 ± 2.16

67.13 ± 0.54 / 21.94 ± 0.39

35.14 ± 1.07 / 51.22 ± 0.84

11.26 ± 1.35 / 31.40 ± 1.40

13.0.0

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

1837

152

32768

False

8,304 ± 1,846 / 1,933 ± 617

2.70

40.89 ± 2.63 / 37.44 ± 2.39

55.33 ± 1.77 / 64.53 ± 0.70

11.23 ± 1.81 / 52.85 ± 2.65

60.69 ± 1.44 / 74.24 ± 0.82

67.23 ± 0.14 / 18.87 ± 0.43

26.84 ± 0.41 / 44.86 ± 0.31

23.89 ± 1.20 / 42.78 ± 0.95

12.5.2

11.0.0

12.1.0

12.5.0

12.1.0

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

1236

128

131328

False

7,436 ± 1,846 / 1,508 ± 479

2.72

56.41 ± 1.79 / 52.05 ± 1.57

59.46 ± 1.16 / 65.61 ± 1.08

8.36 ± 0.71 / 49.50 ± 2.90

47.26 ± 3.09 / 67.95 ± 2.10

66.61 ± 0.54 / 21.06 ± 1.04

25.59 ± 0.93 / 43.75 ± 0.65

18.56 ± 1.15 / 38.13 ± 0.88

13.0.0

AI-Sweden-Models/gpt-sw3-20b-instruct (few-shot)

20918

64

2048

True

1,831 ± 587 / 268 ± 90

2.73

39.21 ± 1.52 / 34.08 ± 1.88

65.58 ± 1.59 / 57.86 ± 1.13

7.82 ± 1.43 / 51.19 ± 1.75

72.25 ± 2.19 / 83.08 ± 1.30

64.07 ± 0.60 / 19.27 ± 0.54

14.98 ± 0.77 / 36.11 ± 0.75

13.55 ± 1.51 / 34.03 ± 1.55

9.3.1

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

3483

49

2048

True

2,732 ± 868 / 662 ± 238

2.75

60.64 ± 2.11 / 55.14 ± 2.01

61.20 ± 1.16 / 61.92 ± 1.68

7.63 ± 2.79 / 46.39 ± 3.79

69.83 ± 2.12 / 80.15 ± 1.27

56.62 ± 3.70 / 13.80 ± 2.77

16.29 ± 0.76 / 36.25 ± 0.61

10.37 ± 1.17 / 32.60 ± 0.89

13.0.0

tiiuae/Falcon3-1B-Instruct (few-shot)

1669

131

8192

True

9,270 ± 2,690 / 1,434 ± 437

2.75

43.00 ± 1.94 / 39.96 ± 1.87

54.47 ± 1.37 / 64.39 ± 0.56

17.44 ± 1.67 / 50.79 ± 2.44

53.15 ± 1.38 / 70.06 ± 0.79

67.33 ± 0.15 / 22.37 ± 0.27

20.49 ± 1.00 / 38.88 ± 0.87

17.50 ± 0.84 / 36.85 ± 0.65

14.1.2

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

1335

49

4096

True

7,964 ± 2,255 / 1,299 ± 433

2.79

46.26 ± 3.26 / 40.49 ± 2.54

63.47 ± 0.73 / 60.77 ± 1.39

13.17 ± 1.60 / 54.31 ± 1.93

59.32 ± 1.48 / 72.97 ± 0.96

66.43 ± 0.60 / 20.68 ± 0.92

9.68 ± 1.07 / 31.43 ± 0.85

4.99 ± 0.71 / 26.87 ± 0.63

13.2.0

TrustLLMeu/baseline-7-8b_1t-tokens_llama (few-shot)

7800

100

4096

True

6,197 ± 1,118 / 1,730 ± 577

2.84

45.89 ± 2.63 / 42.96 ± 2.56

59.29 ± 1.35 / 65.59 ± 1.02

9.11 ± 2.36 / 52.02 ± 2.16

66.74 ± 1.21 / 77.75 ± 1.09

68.17 ± 0.69 / 22.50 ± 0.73

0.92 ± 1.20 / 23.00 ± 0.81

0.61 ± 1.04 / 25.00 ± 0.60

13.0.0

google/gemma-2b (few-shot)

2506

256

8192

True

6,087 ± 1,046 / 1,902 ± 563

2.89

19.65 ± 5.96 / 18.64 ± 5.49

62.14 ± 1.16 / 67.81 ± 0.65

8.30 ± 1.63 / 45.01 ± 3.82

66.30 ± 1.42 / 77.75 ± 0.63

66.51 ± 0.85 / 21.76 ± 0.84

20.38 ± 0.82 / 38.44 ± 0.70

7.41 ± 1.01 / 29.08 ± 0.61

12.5.2

12.1.0

VAGOsolutions/SauerkrautLM-Gemma-2b (few-shot)

2506

256

8192

False

3,607 ± 565 / 1,212 ± 349

2.91

23.28 ± 6.45 / 21.81 ± 5.61

61.91 ± 2.08 / 67.80 ± 1.12

6.92 ± 2.36 / 44.29 ± 3.87

64.68 ± 1.06 / 76.60 ± 0.51

65.15 ± 0.86 / 20.15 ± 0.50

22.65 ± 0.87 / 41.40 ± 0.67

14.54 ± 1.87 / 34.39 ± 1.49

12.6.1

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

2506

256

8192

False

6,471 ± 1,142 / 1,961 ± 584

2.92

40.05 ± 2.56 / 33.77 ± 1.94

48.83 ± 1.00 / 60.88 ± 0.70

5.83 ± 1.52 / 50.74 ± 1.73

63.77 ± 1.40 / 76.59 ± 0.77

67.28 ± 0.27 / 22.84 ± 0.41

18.21 ± 0.70 / 37.61 ± 0.55

10.84 ± 1.23 / 31.84 ± 1.00

12.5.2

12.1.0

12.4.0

12.1.0

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

1657

128

4096

True

15,009 ± 4,072 / 2,702 ± 878

2.97

37.47 ± 1.64 / 32.39 ± 2.53

58.61 ± 2.46 / 62.78 ± 1.22

5.30 ± 1.43 / 46.95 ± 3.51

63.26 ± 0.73 / 74.41 ± 0.65

67.24 ± 0.36 / 21.22 ± 0.36

3.93 ± 0.84 / 28.88 ± 0.73

1.26 ± 1.14 / 25.56 ± 0.82

13.1.0

LumiOpen/Viking-13B (few-shot)

14030

131

4352

True

840 ± 79 / 400 ± 124

3.01

42.78 ± 4.24 / 40.64 ± 2.84

59.90 ± 4.98 / 54.05 ± 2.70

5.68 ± 1.91 / 50.82 ± 2.18

58.52 ± 1.88 / 68.97 ± 1.86

64.83 ± 0.95 / 18.37 ± 0.46

1.63 ± 0.89 / 23.51 ± 0.83

0.54 ± 1.01 / 24.93 ± 0.46

12.10.5

openGPT-X/Teuken-7B-instruct-research-v0.4 (few-shot)

7453

251

4096

False

1,254 ± 328 / 243 ± 83

3.05

50.73 ± 2.64 / 38.64 ± 1.60

27.52 ± 3.38 / 31.81 ± 3.98

2.96 ± 2.64 / 35.23 ± 1.82

63.42 ± 1.34 / 76.38 ± 1.18

68.74 ± 0.08 / 21.51 ± 0.20

16.65 ± 0.64 / 35.86 ± 0.47

10.69 ± 1.12 / 29.62 ± 0.77

14.0.4

NorwAI/NorwAI-Llama2-7B (few-shot)

7033

68

4224

True

4,438 ± 1,128 / 1,028 ± 346

3.07

38.51 ± 3.33 / 38.08 ± 2.64

63.60 ± 2.87 / 58.50 ± 1.25

2.23 ± 1.32 / 34.15 ± 0.60

45.44 ± 3.61 / 59.52 ± 2.98

67.11 ± 0.90 / 21.77 ± 0.72

5.02 ± 1.22 / 25.76 ± 1.01

2.08 ± 0.75 / 25.95 ± 0.35

12.10.4

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

620

152

32768

True

11,371 ± 2,924 / 2,122 ± 692

3.07

37.51 ± 1.56 / 33.24 ± 2.24

57.15 ± 2.35 / 52.82 ± 1.40

2.94 ± 2.00 / 44.53 ± 3.65

42.57 ± 1.97 / 55.17 ± 1.29

65.22 ± 0.29 / 19.22 ± 0.18

18.24 ± 1.14 / 37.15 ± 0.82

10.89 ± 1.01 / 32.65 ± 0.85

12.5.2

10.0.1

12.1.0

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

620

152

32768

False

11,740 ± 3,000 / 2,209 ± 721

3.11

33.86 ± 2.16 / 32.80 ± 2.21

55.41 ± 2.17 / 54.48 ± 1.65

1.15 ± 1.81 / 34.47 ± 1.12

53.34 ± 1.21 / 64.09 ± 1.26

65.81 ± 1.90 / 20.88 ± 0.66

11.66 ± 1.13 / 32.84 ± 0.97

5.22 ± 1.17 / 27.95 ± 1.10

12.5.2

11.0.0

12.1.0

12.5.0

12.1.0

NbAiLab/nb-llama-3.1-8B-Instruct (few-shot)

8030

128

131072

True

1,296 ± 335 / 246 ± 84

3.12

40.91 ± 6.01 / 37.55 ± 4.86

47.12 ± 10.26 / 54.09 ± 8.04

6.03 ± 1.71 / 49.55 ± 3.61

51.34 ± 2.37 / 65.28 ± 2.29

66.65 ± 1.14 / 20.92 ± 1.64

4.91 ± 2.84 / 25.92 ± 1.81

1.21 ± 1.04 / 25.67 ± 0.89

14.0.4

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

1385

49

4096

True

7,808 ± 2,183 / 1,289 ± 428

3.12

29.84 ± 4.45 / 28.13 ± 3.61

64.13 ± 1.51 / 56.76 ± 0.96

3.99 ± 1.63 / 43.68 ± 3.91

55.74 ± 1.72 / 65.95 ± 1.64

64.55 ± 0.62 / 19.04 ± 0.65

1.42 ± 1.20 / 26.10 ± 0.80

1.42 ± 0.61 / 25.58 ± 0.70

13.2.0

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

1236

128

131328

True

7,577 ± 1,884 / 1,555 ± 492

3.12

35.08 ± 5.88 / 32.44 ± 4.89

54.40 ± 2.92 / 64.38 ± 2.10

2.97 ± 0.84 / 45.05 ± 4.18

58.30 ± 2.84 / 71.04 ± 1.83

62.19 ± 1.70 / 16.53 ± 1.13

7.91 ± 0.72 / 29.30 ± 0.46

1.60 ± 1.21 / 26.25 ± 0.85

13.0.0

openGPT-X/Teuken-7B-instruct-commercial-v0.4 (few-shot)

7453

251

4096

True

1,438 ± 410 / 233 ± 79

3.13

44.48 ± 3.17 / 36.31 ± 2.23

23.69 ± 3.36 / 25.98 ± 3.59

8.52 ± 2.60 / 51.57 ± 2.62

56.97 ± 1.53 / 71.38 ± 1.28

68.81 ± 0.06 / 22.48 ± 0.20

15.31 ± 1.45 / 35.07 ± 1.13

9.35 ± 1.36 / 29.47 ± 1.09

14.0.4

NorwAI/NorwAI-Mistral-7B (few-shot)

7537

68

4224

True

3,035 ± 503 / 911 ± 300

3.14

35.84 ± 5.12 / 35.76 ± 4.29

56.87 ± 8.16 / 59.62 ± 5.61

3.08 ± 2.98 / 40.66 ± 4.63

52.77 ± 1.53 / 66.64 ± 1.58

63.96 ± 1.61 / 19.20 ± 1.54

7.42 ± 1.60 / 28.95 ± 1.40

2.24 ± 1.08 / 25.72 ± 0.82

12.10.4

AI-Sweden-Models/gpt-sw3-20b (few-shot)

20918

64

2048

True

1,875 ± 673 / 261 ± 91

3.21

45.86 ± 3.18 / 40.23 ± 2.41

62.08 ± 3.29 / 55.11 ± 1.68

6.62 ± 2.43 / 48.79 ± 3.77

65.29 ± 1.81 / 77.71 ± 0.98

43.45 ± 7.12 / 15.14 ± 2.47

9.10 ± 1.33 / 30.31 ± 0.94

8.35 ± 0.79 / 29.79 ± 0.76

9.3.1

allenai/OLMo-7B (few-shot)

6888

50

2304

True

5,403 ± 1,133 / 1,294 ± 423

3.27

38.23 ± 3.10 / 36.38 ± 2.60

60.70 ± 1.80 / 66.82 ± 1.12

-0.19 ± 2.28 / 38.77 ± 3.32

61.93 ± 1.98 / 73.71 ± 1.57

51.32 ± 0.91 / 11.96 ± 0.57

5.00 ± 1.53 / 27.59 ± 1.14

1.10 ± 1.02 / 25.71 ± 0.66

12.5.2

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

2768

50

33024

True

2,722 ± 495 / 766 ± 250

3.31

28.63 ± 4.74 / 27.07 ± 4.13

66.55 ± 0.72 / 58.18 ± 0.62

1.47 ± 1.57 / 45.89 ± 2.92

35.00 ± 2.01 / 47.83 ± 2.25

62.05 ± 1.63 / 16.80 ± 1.01

-0.41 ± 1.21 / 23.76 ± 0.63

-0.04 ± 0.77 / 25.03 ± 0.52

13.0.0

HuggingFaceTB/SmolLM2-360M (few-shot)

362

49

8192

True

22,023 ± 6,203 / 3,675 ± 1,231

3.36

31.14 ± 1.79 / 28.54 ± 0.86

43.97 ± 5.28 / 55.08 ± 4.26

3.49 ± 2.49 / 46.52 ± 4.13

47.91 ± 4.97 / 60.41 ± 3.91

62.20 ± 1.04 / 17.61 ± 0.50

0.12 ± 1.55 / 23.00 ± 0.82

0.13 ± 1.34 / 24.53 ± 0.49

13.1.0

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

362

49

8192

True

21,777 ± 6,115 / 3,617 ± 1,211

3.37

30.73 ± 4.30 / 29.47 ± 4.10

59.51 ± 3.73 / 54.82 ± 2.43

1.55 ± 1.90 / 43.18 ± 5.08

49.03 ± 1.47 / 60.00 ± 1.53

57.73 ± 4.93 / 15.68 ± 1.21

0.11 ± 1.23 / 23.20 ± 0.56

-0.06 ± 0.39 / 24.60 ± 0.44

13.1.0

allenai/OLMo-7B-Twin-2T (few-shot)

6888

50

2304

True

5,484 ± 1,125 / 1,317 ± 425

3.49

25.36 ± 8.05 / 24.05 ± 7.34

56.91 ± 2.41 / 66.15 ± 1.54

7.10 ± 2.89 / 49.36 ± 4.03

58.60 ± 5.37 / 70.37 ± 4.71

46.16 ± 1.28 / 11.33 ± 0.70

0.71 ± 0.85 / 24.97 ± 0.62

0.82 ± 1.01 / 25.62 ± 0.74

12.5.2

allenai/OLMo-1B (few-shot)

1177

50

2304

True

8,536 ± 1,926 / 1,940 ± 619

3.62

26.47 ± 6.25 / 28.27 ± 5.35

60.05 ± 3.94 / 56.18 ± 1.90

0.72 ± 1.90 / 42.84 ± 3.50

43.87 ± 2.49 / 55.59 ± 2.44

46.18 ± 1.20 / 10.06 ± 0.67

-0.87 ± 0.99 / 23.08 ± 0.63

0.20 ± 1.12 / 24.94 ± 0.86

12.5.2

12.1.0

PleIAs/Pleias-1.2b-Preview (few-shot)

1195

66

2048

True

10,756 ± 3,589 / 1,157 ± 670

3.69

40.45 ± 3.27 / 37.90 ± 2.99

47.89 ± 3.76 / 56.99 ± 2.11

0.28 ± 1.01 / 44.40 ± 3.07

26.77 ± 5.45 / 35.60 ± 6.25

50.31 ± 7.81 / 13.58 ± 2.12

1.03 ± 0.82 / 26.95 ± 0.56

-0.30 ± 1.30 / 24.99 ± 1.10

14.0.4

14.1.2

14.0.4

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

unknown

32

4096

False

10,890 ± 2,686 / 2,186 ± 750

3.79

18.69 ± 7.23 / 18.50 ± 6.51

21.95 ± 6.30 / 33.38 ± 4.79

0.01 ± 1.91 / 39.40 ± 3.94

36.51 ± 2.07 / 50.66 ± 1.90

60.11 ± 1.39 / 16.29 ± 0.69

3.71 ± 0.75 / 28.35 ± 0.99

0.62 ± 0.99 / 24.68 ± 0.55

9.3.1

12.5.2

9.3.1

PleIAs/Pleias-Nano (few-shot)

1195

66

2048

True

2,519 ± 841 / 323 ± 104

3.81

21.60 ± 4.23 / 23.22 ± 3.33

45.04 ± 5.19 / 50.60 ± 4.14

-0.46 ± 1.47 / 44.56 ± 3.38

33.46 ± 2.84 / 44.25 ± 3.38

49.12 ± 7.39 / 12.74 ± 2.01

2.37 ± 0.75 / 27.67 ± 0.77

-0.33 ± 1.21 / 25.00 ± 1.15

14.0.4

14.1.2

14.0.4

PleIAs/Pleias-3b-Preview (few-shot)

3212

66

4096

True

6,513 ± 1,241 / 1,282 ± 644

3.85

27.37 ± 4.57 / 26.50 ± 4.52

36.35 ± 7.92 / 45.58 ± 8.18

-0.37 ± 1.89 / 44.67 ± 2.86

7.42 ± 2.13 / 16.44 ± 2.71

60.89 ± 0.87 / 17.36 ± 0.62

2.21 ± 1.17 / 26.40 ± 1.12

-0.90 ± 0.79 / 24.52 ± 0.70

14.0.4

14.1.2

14.0.4

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

135

49

8192

True

25,602 ± 7,583 / 3,953 ± 1,325

4.00

29.96 ± 3.19 / 28.98 ± 3.29

18.64 ± 8.52 / 28.83 ± 5.86

1.85 ± 1.20 / 44.03 ± 3.98

26.90 ± 1.56 / 36.91 ± 1.53

52.96 ± 1.38 / 14.23 ± 0.78

1.34 ± 0.74 / 25.67 ± 1.03

0.10 ± 1.29 / 24.71 ± 0.69

13.1.0

HuggingFaceTB/SmolLM2-135M (few-shot)

135

49

8192

True

26,346 ± 7,812 / 4,082 ± 1,372

4.05

31.26 ± 3.84 / 30.44 ± 3.28

26.69 ± 10.82 / 34.46 ± 8.00

1.78 ± 1.67 / 43.50 ± 3.99

13.88 ± 1.36 / 22.49 ± 2.14

52.05 ± 1.59 / 12.74 ± 1.04

1.51 ± 0.81 / 25.05 ± 0.75

-0.76 ± 0.97 / 24.52 ± 0.59

13.1.0

NbAiLab/nb-llama-3.2-3B (few-shot)

3213

128

131072

True

1,880 ± 556 / 280 ± 95

4.11

0.02 ± 0.03 / 0.01 ± 0.03

60.98 ± 1.52 / 58.15 ± 0.97

0.00 ± 0.00 / 33.36 ± 0.27

9.94 ± 1.53 / 26.50 ± 2.35

41.10 ± 2.34 / 6.64 ± 0.65

12.17 ± 1.07 / 31.75 ± 1.03

0.62 ± 1.44 / 25.00 ± 0.88

14.0.4

14.1.2

14.0.4

NbAiLab/nb-llama-3.2-1B (few-shot)

1236

128

131072

True

3,424 ± 1,080 / 464 ± 158

4.19

3.98 ± 3.16 / 3.68 ± 2.99

39.54 ± 6.40 / 48.55 ± 7.75

3.41 ± 1.85 / 41.04 ± 4.56

26.96 ± 2.30 / 42.51 ± 3.06

43.27 ± 1.15 / 7.42 ± 0.80

2.45 ± 1.06 / 28.00 ± 0.77

-0.84 ± 1.17 / 24.50 ± 0.79

14.0.4

14.1.2

14.0.4

PleIAs/Pleias-350m-Preview (few-shot)

353

66

2048

True

10,242 ± 3,432 / 1,335 ± 484

4.22

31.79 ± 3.88 / 31.32 ± 2.81

19.13 ± 9.92 / 33.51 ± 6.97

-0.03 ± 1.07 / 36.37 ± 2.34

12.35 ± 1.80 / 21.93 ± 1.63

46.13 ± 6.88 / 11.20 ± 1.59

-1.20 ± 0.68 / 23.43 ± 0.94

0.47 ± 1.47 / 25.42 ± 1.15

14.0.4

14.1.2

14.0.4

PleIAs/Pleias-Pico (few-shot)

353

66

2048

True

2,331 ± 787 / 301 ± 97

4.22

27.45 ± 4.13 / 27.11 ± 3.47

27.39 ± 8.29 / 39.38 ± 6.65

0.31 ± 1.47 / 39.37 ± 3.75

15.62 ± 2.29 / 26.76 ± 1.43

43.83 ± 6.89 / 8.96 ± 1.14

0.65 ± 0.70 / 24.30 ± 0.79

-0.44 ± 1.22 / 24.71 ± 0.98

14.0.4

14.1.2

14.0.4

RJuro/kanelsnegl-v0.1 (few-shot)

7242

32

512

True

5,847 ± 1,029 / 1,640 ± 525

4.53

0.00 ± 0.00 / 0.00 ± 0.00

0.00 ± 0.00 / 19.61 ± 0.22

0.41 ± 0.55 / 33.46 ± 0.37

0.00 ± 0.00 / 3.68 ± 0.43

61.26 ± 0.09 / 6.73 ± 0.06

0.00 ± 0.00 / 22.69 ± 0.63

0.36 ± 0.47 / 24.60 ± 0.43

9.3.1

12.5.1

9.3.1

NorwAI/NorwAI-Mistral-7B-pretrain (few-shot)

7537

68

4224

True

3,024 ± 496 / 909 ± 301

4.66

12.34 ± 2.70 / 12.41 ± 2.54

-1.48 ± 3.09 / 21.17 ± 2.22

-0.48 ± 1.52 / 42.45 ± 3.99

0.72 ± 0.25 / 6.31 ± 0.51

49.61 ± 1.17 / 9.59 ± 0.65

-0.12 ± 0.65 / 22.69 ± 0.62

-0.01 ± 0.01 / 24.59 ± 0.43

12.10.4

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

5465

65

8192

True

7,692 ± 1,423 / 1,960 ± 644

4.81

0.00 ± 0.00 / 0.00 ± 0.00

0.00 ± 0.00 / 19.61 ± 0.22

2.48 ± 1.94 / 34.52 ± 0.85

0.01 ± 0.02 / 0.03 ± 0.02

44.80 ± 0.36 / 0.32 ± 0.04

-0.69 ± 0.77 / 22.76 ± 0.40

0.37 ± 0.95 / 24.71 ± 0.44

13.0.0

ai-forever/mGPT (few-shot)

unknown

100

1024

True

11,734 ± 3,124 / 2,174 ± 720

4.91

1.55 ± 1.98 / 1.45 ± 1.82

3.71 ± 3.16 / 22.09 ± 2.08

-0.42 ± 1.56 / 40.58 ± 3.74

5.58 ± 3.10 / 11.12 ± 4.66

34.62 ± 0.17 / 2.18 ± 0.04

0.37 ± 0.93 / 24.18 ± 0.90

-0.17 ± 0.50 / 24.82 ± 0.41

9.3.1

10.0.1

11.0.0

12.5.1

12.0.0

11.0.0