fedi-index: Language: Français

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

12

A

1,454 ±19

92

+105◆

A

14

A

1,451 ±12

90

+80

A

17

A

1,448 ±17

88

+44

A

22

A

1,447 ±22

84

+60

A

25

A

1,440 ±14

82

+45

A

27

A

1,438 ±13

81

+36

A

31

B

1,435 ±12

78

+33

A

35

B

1,434 ±14

75

+44

A

40

B

1,428 ±23

71

+33

A

41

B

1,427 ±11

70

+27

A

42

B

1,426 ±21

69

+46

A

47

B

1,421 ±19

66

+26

A

51

B

1,420 ±11

63

+36

A

54

B

1,419 ±16

60

+27

A

55

B

1,418 ±17

60

+23

A

60

B

1,414 ±25

56

+68

A

62

B

1,413 ±13

54

+75

A

63

B

1,413 ±13

54

+50

A

67

B

1,408 ±24

51

-19

A

69

C

1,407 ±17

49

+14

A

70

C

1,407 ±13

48

+71

A

77

C

1,396 ±18

43

+38

A

80

C

1,393 ±11

41

+32

A

81

C

1,392 ±12

40

+50

A

82

C

1,391 ±27

40

+20

A

83

C

1,391 ±25

39

+55◆

A

84

C

1,385 ±26

38

+18

A

87

C

1,382 ±18

36

+11

A

88

C

1,382 ±26

35

+25

A

90

C

1,379 ±18

34

+14

A

91

C

1,379 ±25

33

+32

A

93

C

1,372 ±23

31

+18

A

97

C

1,368 ±23

28

-2

A

98

C

1,364 ±25

28

+6

A

99

C

1,362 ±25

27

-22

A

106

D

1,343 ±23

22

-2

A

107

D

1,343 ±23

21

-36

A

109

D

1,342 ±17

19

-30

A

111

D

1,341 ±15

18

+9

A

115

D

1,326 ±25

15

-29

A

116

D

1,309 ±27

14

—

A

123

D

1,174 ±22

9

-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