fedi-index: Language: Deutsch

Best models for this language by Arena votes in it. Provisional: one source family.

Live data for 2026-10 (recalculated daily) · Methodology v1 · Methodology · History · JSON

All Open weights Closed

#

Tier

Model

fedi-index

Percentile

Value

Sources

14

A

1,445 ±20

91

+44

A

22

A

1,438 ±17

85

+45

A

23

A

1,437 ±16

85

+36

A

28

A

1,432 ±15

81

+80

A

30

B

1,432 ±15

80

+33

A

33

B

1,428 ±23

78

+105◆

A

36

B

1,426 ±14

76

+27

A

37

B

1,425 ±17

75

+44

A

40

B

1,423 ±18

73

+26

A

43

B

1,420 ±14

71

+36

A

44

B

1,420 ±20

70

+26

A

46

B

1,420 ±20

68

+23

A

56

B

1,407 ±16

62

+50

A

59

B

1,403 ±22

59

+14

A

60

B

1,402 ±17

59

+75

A

62

B

1,401 ±17

57

+71

A

66

B

1,397 ±21

54

+32

A

70

B

1,391 ±23

52

+19

A

71

B

1,391 ±15

51

+50

A

72

B

1,389 ±20

50

+18

A

74

C

1,387 ±22

49

+38

A

75

C

1,384 ±14

48

+32

A

83

C

1,378 ±21

43

+14

A

88

C

1,368 ±20

39

+25

A

89

C

1,367 ±17

38

-2

A

92

C

1,358 ±20

36

+11

A

94

C

1,354 ±19

35

+9

A

95

C

1,349 ±18

34

-2

A

96

C

1,349 ±22

34

-30

A

99

C

1,346 ±16

32

-36

A

101

C

1,337 ±23

30

-14

A

104

C

1,325 ±20

28

-22

A

105

C

1,316 ±17

27

—

A

108

C

1,313 ±17

25

-29

A

125

D

1,181 ±14

13

-236

A

AArena (human votes) ◆ Pareto frontier: nothing is both better and cheaper

fedi-index is an Elo-equivalent on a fixed scale: comparable between months within one methodology version. Tiers S–D: by percentile in the slice, and no better than the confidence interval allows (cut-offs in the methodology). This slice has one source family, so all its scores are provisional. Methodology